Optimize the Harness, Not the Model: Fitting a Minimal Agent to a Self-Hostable LLM for Investigative Analysis
arXiv |
2026-11 |
— |
— |
Can We Stop the Backtest Yet? Anytime-Valid Promotion Tests for Adaptive Financial Research Agents
arXiv |
2026-11 |
0 |
— |
ScienceCUA: Resource-to-Skill Construction and Execution-Guided Skill Evolution for Scientific Computer Use Agents
arXiv |
2026-11 |
— |
— |
Auto-Research Agent on MSA Organization to Improve Protein-Protein Complex Structure Prediction
arXiv |
2026-11 |
— |
— |
CARE: Grounded Recursive Self-Improvement for Production Customer-Service Agents
arXiv |
2026-11 |
— |
— |
GeoGuesser Bench: Benchmarking Vision-Language Agents on Interactive Geolocation
arXiv |
2026-11 |
— |
— |
Strategist: Evolving Reasoning with Agents
arXiv |
2026-11 |
— |
— |
DiabAgent: Toward Agentic Physiological Control
arXiv |
2026-11 |
— |
— |
FedReady: Bounded Agents for Federation Formation
arXiv |
2026-11 |
— |
— |
An Agentic Framework for Automated Power System Studies and Grid Connection Assessment
arXiv |
2026-11 |
— |
— |
Top-$k$ Retrieval Leaves Memory Value Undetermined
arXiv |
2026-11 |
— |
— |
CoworkOS:Scaling Agent Coworkers for Real-world Automation
arXiv |
2026-11 |
— |
— |
ComplianceAtlas: A Taxonomy-Driven Agentic AI Framework for Automated AI Act Gap Analysis and Remediation
arXiv |
2026-11 |
— |
— |
Action Verification through Conjecture and Refutation for Language Model Agents
arXiv |
2026-11 |
— |
— |
RxnMiner: A Multimodal Agent for Structured Reaction Data Extraction from Chemistry Figures
arXiv |
2026-11 |
— |
GitHub · ★ 4 |
Predictive Safety Guardrails Do Not Always Make LLM Agents Safer
arXiv |
2026-11 |
— |
— |
Cumulative and Audited Guardrail Tool Evolution for LLM Agent Safety
arXiv |
2026-11 |
— |
— |
MAGAR: Multimodal Agents for GUI-Based 3D Mechanical Assembly and Minimal Rework
arXiv |
2026-11 |
— |
— |
RecGroupMemory: Feedback-Aware Cohort Memory for E-Commerce Recommendation
arXiv |
2026-11 |
— |
— |
LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents
arXiv |
2026-11 |
— |
— |
Can Coding Agents Sustain Continual Evolution without Forgetting?
arXiv |
2026-11 |
— |
GitHub · ★ 0 |
Math Foundation Formalization Agent: A General-Purpose Approach to Formalizing Mathematical Foundations
arXiv |
2026-11 |
— |
— |
Discovering hierarchical governing equations for community-structured complex network dynamics with LLM agents
arXiv |
2026-11 |
— |
— |
IssueLighter: An Issue Type-Aware LLM Agent for Issue Report Validity Assessment
arXiv |
2026-11 |
— |
— |
Scalaris: A Scalable Enterprise Memory Architecture for Conversational AI Assistants
arXiv |
2026-11 |
— |
— |
StateFence: Task-State Protection Against Evidence-Chain Contamination in Blue-Team LLM Agents
arXiv |
2026-11 |
— |
— |
Chain on Demand: Deferred Evidence Organization for Long-Term Conversational Memory
arXiv |
2026-11 |
— |
— |
Late Autumn: LLM Agents Specialize a General Multi-Robot Planner for One Warehouse
arXiv |
2026-11 |
— |
— |
AgiWorld: An Agentic Video World Model for Building Persistent Worlds
arXiv |
2026-11 |
— |
— |
MemShift: Separating Detection from Repair in LLM-Derived Personal Memory
arXiv |
2026-11 |
— |
GitHub · ★ 1 |
VeriHarm-Bench: Measuring Harmful Success in LLM Coding Agents Verifiably
arXiv |
2026-11 |
— |
— |
ACTOR: Action-Centered Temporal Organization for Literary Role-Playing Agents
arXiv |
2026-11 |
— |
GitHub · ★ 5 |
A State Change Is Local, Its Consequences Are Not: Dependency-Aware State Revision for Long-Term Agent Memory
arXiv |
2026-11 |
— |
— |
ParetoSaddler: Pareto-Guided Optimization of Agent Harnesses for Quality and Cost
arXiv |
2026-11 |
— |
Project |
Self-Evolving Agentic Workflows via Semantic Decomposition of Complex Tasks
arXiv |
2026-11 |
— |
GitHub · ★ 0 |
CONFIRM: Governing Scientific Reporting for Trustworthy Agentic Neuroimaging
arXiv |
2026-11 |
— |
— |
A Self-Exploring and Self-Evolving Paradigm for Adaptive Edge AI Development
arXiv |
2026-11 |
— |
— |
Towards Self-Evolving Narrative Systems: Real-Time Adaptation and Long-Term Optimization with LLMs
arXiv |
2026-11 |
— |
— |
Latent Harness: Compiling Agent Behaviors into Post-Block Activation Shifts
arXiv |
2026-11 |
— |
— |
Decoupled Predictive Decision Making: A Shared Conditional Predictor for Agents
arXiv |
2026-11 |
— |
— |
NEIGHBOR: Aligning Service Behaviour and Evidence Use in an LLM Agent for Community Resident Services
arXiv |
2026-11 |
— |
— |
Delivered, Not Used: Evaluating Agent Memory at Its Point of Use
arXiv |
2026-11 |
— |
— |
Just-in-Time Skill Retrieval via Long-Horizon State Tracking
arXiv |
2026-11 |
— |
— |
MemHarmBench: Evaluating Cross-Session Prompt Injection in Memory-Persistent AI Agents
arXiv |
2026-11 |
— |
GitHub · ★ 3 |
\textsc{ChatDify}: Automating End-to-End Dify Workflow Synthesis from Natural Language via Multi-Agent Collaboration
arXiv |
2026-11 |
— |
— |
I-PTC: Interactive Programmatic Tool Calling for Stateful Tool-Augmented Agents
arXiv |
2026-11 |
— |
— |
Learning by Doing: Autonomously Compiling and Curating Agent Skills from Verified Traces
arXiv |
2026-11 |
— |
— |
Qubrio: High-Performance Quantum Compilation via Multi-Agent LLM Collaboration
arXiv |
2026-11 |
— |
GitHub · ★ 2 |
Beyond Prompt Injection: The Tool Update Gap in LLM Agent Security
arXiv |
2026-11 |
— |
— |
Step-Zero Attack: Agent Safety Before Observability
arXiv |
2026-11 |
— |
— |
RECEIPT: Measuring and Repairing the Return Assumption of LLM Agents in Web Service Operations
arXiv |
2026-11 |
— |
— |
AgenticMobility: An Open-Source Platform for Multi-AI-Agent Mobility Dynamics
arXiv |
2026-11 |
— |
— |
MemRole: From Memories to Decisions for Long-Horizon Reasoning in Role-Playing Agents
arXiv |
2026-11 |
— |
— |
Evolving Memory, Learning to Meta-Control: Test-Time Adaptation for Experience-augmented LLM Agents
arXiv |
2026-11 |
— |
— |
Principled Human-AI Collaboration in Agentic Workflows
arXiv |
2026-11 |
— |
— |
An auditable agentic pipeline for variant-to-gene prioritisation, validated against orthogonal functional evidence
arXiv |
2026-11 |
— |
— |
Can Large Language Model Agents Learn by Watching Tutorial Videos?
arXiv |
2026-11 |
— |
— |
ERIA: An Instruction-Optimized Agent for Clinical EEG Reporting
arXiv |
2026-11 |
— |
— |
Decision Models as Policy Gates for Coding Agents: Explicit Policy Boundaries Improve Detection and Reduce False Alarms
arXiv |
2026-11 |
— |
GitHub · ★ 1 |
Multi-Agent Collusion and the J-Space
arXiv |
2026-11 |
— |
— |
Can Large Language Models Extend Their Agentic Capabilities to Goal Pursuit in Open-Ended Multi-Turn Interaction?
arXiv |
2026-11 |
— |
— |
When Should an Agent Refresh? Adaptive Index Reuse for Efficient Long-Horizon Reasoning
arXiv |
2026-11 |
— |
— |
Designing Informative Actions: Action Construction for Long-Horizon Research Agents
arXiv |
2026-10 |
— |
— |
Auriga Technical Report: A Task-Agnostic Agent Harness for Zero-Shot Embodied Manipulation
arXiv |
2026-10 |
— |
GitHub · ★ 22 |
When LLMs Modify Themselves: Convergence, Values, and Stopping in Iterative Self-Improvement
arXiv |
2026-10 |
— |
— |
CivicBench: A Benchmark and Memory Framework for Web-Based Civic Data Analytics
arXiv |
2026-10 |
— |
— |
Constraint Dependency Graph Retrieval for Compliance Detection on Long SKILL Documents
arXiv |
2026-10 |
— |
— |
GURU: A Hierarchical Neuro-Symbolic Multi-Agent System for Verification-Aware Automation of Mission-Critical Digital Engineering and Operations
arXiv |
2026-10 |
— |
— |
Can Agentic AI Undermine Its Own Assurance Loop? Closed-Loop Dependency Assurance for Cognitive Network Digital Twins
arXiv |
2026-10 |
— |
— |
Who Schedules Agent Memory? A Sidecar Approach to Long-Horizon Tasks
arXiv |
2026-10 |
— |
— |
FastAgent: A Physics-Aware FastFluent-to-Fluent Workflow for Rapid CFD Screening and Industrial Steam Dewaxing Design
arXiv |
2026-10 |
— |
— |
Autonomous Analysis of Thermoelectric Materials Big Data by Multi-Agentic AI and Construction of a Knowledge Base and Specialized AI
arXiv |
2026-10 |
— |
— |
Skill++: Towards Lifelong Agents via Routing-Aware Skill Refinement
arXiv |
2026-10 |
— |
— |
Agentic Context Management for Multi-Turn Human-AI Conversations
arXiv |
2026-10 |
— |
— |
The Verifier Is Part of the Simulator: Evidence-View Gaming in Agentic Equation Discovery
arXiv |
2026-10 |
— |
GitHub · ★ 1 |
Toward Bounding Coding-Agent Autonomy in Software Maintenance: A Pilot Study of Boundary Contracts
arXiv |
2026-10 |
— |
— |
Dependable Agents Need Hard and Fuzzy Rules
arXiv |
2026-10 |
— |
— |
Illusion: Safeguarding Topology Confidentiality in LLM Multi-Agent Systems via Behavior-Preserving Obfuscation
arXiv |
2026-10 |
— |
— |
Agent4FreeFem: A Verification-Guided Framework for Autonomous Topology Optimization
arXiv |
2026-10 |
— |
— |
Old Is Not False: Valid-Time Memory for Interactive Video World Models
arXiv |
2026-10 |
— |
— |
Getting a Grip: Operating Human-Agent Software Teams Through Governed Messages
arXiv |
2026-10 |
— |
— |
Discovering Models for Experimental Quantum Systems with Agentic AI
arXiv |
2026-10 |
— |
— |
CompSkill: Compiling Execution Experience for Skill Retrieval
arXiv |
2026-10 |
— |
GitHub · ★ 0 |
When 200 OK Is Not an Answer: Silent Fetch Failures in Agentic Web Content Acquisition
arXiv |
2026-10 |
— |
— |
Exo: An Architecture for Self-Improving Systems at Runtime
arXiv |
2026-10 |
— |
— |
MemU-Bench: Benchmarking GUI Agents on Long-Horizon User Interactions
arXiv |
2026-10 |
— |
— |
Agent Stack for ML Training Lifecycle: Generation, Validation, and Optimization
arXiv |
2026-10 |
— |
— |
TradeWar: A Benchmark for Execution-Grounded Trade Negotiation with Adaptive LLM Agents
arXiv |
2026-10 |
— |
— |
EvoMCP: Co-evolving MCP agents and training data
arXiv |
2026-10 |
— |
GitHub · ★ 0 |
From 1D Benchmarks to Complex 3D Geometries: Can Agents Reliably Specify Problems for Automated PDE Simulation?
arXiv |
2026-10 |
— |
— |
MEMIG: Self-evolving Memory Construction for Antibody Optimization
arXiv |
2026-10 |
— |
— |
ANGLER: Verifier-Grounded Benchmarking with an Adaptive Attacker for Web Agents
arXiv |
2026-10 |
— |
— |
Context as Infrastructure: Knowledge Compilation for Retrieval in Agentic Systems with Harnesses
arXiv |
2026-10 |
— |
GitHub · ★ 93,418 |
Dependence-Aware Risk Assessment and Optimal Design of AI-Agent Safeguard Systems
arXiv |
2026-10 |
— |
— |
Taurus: LLM-in-the-Loop CT Annotation Matches Coding Agents at Half the Cost
arXiv |
2026-10 |
— |
— |
Fidelity Collapses with Table Size, Not Turn Count: Delivery Architecture in a Deployed Tool-Using Agent
arXiv |
2026-10 |
— |
— |
SPARK: Self-Correcting Embodied Memory via Physical Re-Observation
arXiv |
2026-10 |
— |
— |
Structure-Aware Skill Retrieval and Arbitration for Optimization Modeling
arXiv |
2026-10 |
— |
— |
Learning to Interact and Direct: Co-Evolving Video Director Agents
arXiv |
2026-10 |
— |
— |
Real Computation or Merely Naming? A Pre-Registered Mechanism Audit of 25 Cognitive-Architecture Modules in an LLM-Based Agent System
arXiv |
2026-10 |
— |
— |
Verify--Backtrack Loop: Structured Precondition Recovery for Long-Horizon Language Agents
arXiv |
2026-10 |
— |
— |
Evidence-Guided Recovery for AI Agents with Ambiguous Tool Outcomes and Partial Execution
arXiv |
2026-10 |
— |
— |
Cross-Interface Authorization for AI Computer-Use Agents
arXiv |
2026-10 |
— |
— |
Ask on Demand: Targeted Peer Communication for Parallel Agentic Systems
arXiv |
2026-10 |
— |
— |
Operon-S1: a fast decision controller for long-horizon biomedical agents
arXiv |
2026-10 |
— |
— |
When Should AI Agents Ask? Clarification Requests Under Limited Human Attention
arXiv |
2026-10 |
— |
— |
RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory
arXiv |
2026-10 |
— |
Project |
KEEPER: Learning Control-Sufficient Textual Memory for Long-Horizon LLM Agents
arXiv |
2026-10 |
— |
— |
When Safety Checks Become Privacy Risks: A Privacy-Preserving Paradigm for LLM Agent Safety Auditing
arXiv |
2026-10 |
— |
— |
ART of PIV: Agentic Real-Time Particle Image Velocimetry for Active Flow Control
arXiv |
2026-10 |
— |
— |
Lineage Is Memory: Parent Inheritance and Feedback Disclosure in Short-Budget LLM-Guided Optimization
arXiv |
2026-10 |
— |
GitHub · ★ 1 |
MEMENTO: Learning Context Management for Long Context Processing Agents
arXiv |
2026-10 |
— |
— |
TraceSpec: Understanding and Improving Speculative Decoding for LLM Agents
arXiv |
2026-10 |
— |
— |
Completeness and Cost of Retrieval Structures for Multi-Tool Requests in LLM Agents
arXiv |
2026-10 |
— |
— |
Hybrid Computer Use: A Unified Programmatic, Semantic, and Visual Interface for General Agents
arXiv |
2026-10 |
— |
— |
Towards Portable Specification of the Agent Harness
arXiv |
2026-10 |
— |
— |
TSFLab: A Self-Improving Agent-Agnostic Workbench for Time Series Forecasting Research
arXiv |
2026-10 |
— |
GitHub · ★ 65 |
MetaVul: Agentic Harness Discovery for Vulnerability Detection
arXiv |
2026-10 |
— |
— |
Collaborative AI agents for autonomous experimental research with solid-state quantum spins
arXiv |
2026-10 |
0 |
GitHub · ★ 12 |
KARMA: Training-Free Knowledge-Atomic Harness Evolution for Industrial Advertising Agents
arXiv |
2026-10 |
0 |
— |
MoLRSI: A Recursive Self-Improving AI agent for Molecular Discovery
arXiv |
2026-10 |
0 |
— |
Letting AI Formulate the Research Question: Autonomous Research Agents
arXiv |
2026-10 |
0 |
— |
From Intent to Improvement: Agentic Dataset Search for Better Model Training
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
Compile Away Uncertainty: Self-Evolving Agent Skills via Programmatic Compilation
arXiv |
2026-10 |
0 |
— |
HEMO: Harness Self-Evolution for Multi-Objective Combinatorial Optimization
arXiv |
2026-10 |
0 |
— |
Sandman: A Sandbox Service for Production Agentic Reinforcement Learning
arXiv |
2026-10 |
— |
— |
Well-Specified and Well-Tested? Assessing SWE-bench-Style Datasets with a Self-Evolving Agent
arXiv |
2026-10 |
0 |
— |
Appreciation Teaches Unauthorized Execution: Self-Evolution Degrades the Robustness of Personal Agents to Indirect Prompt Injection
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
Mage-VL-Cowork: A Full Training Recipe to Learn Hybrid Computer Use From Evolved Harnesses
arXiv |
2026-10 |
— |
Project |
How Specific Do I Need to Be? Effects of User Query Underspecification on Deep Research Agents
arXiv |
2026-10 |
0 |
— |
LanG 2.0: Energy-Efficient Agentic AI for Semi-Automated Security Operations at Scale
arXiv |
2026-10 |
— |
— |
Capability-Grounded Agentic Planning
arXiv |
2026-10 |
— |
— |
Agentic World Modeling for Agent Test-time Scaling without Environment Replica
arXiv |
2026-10 |
— |
— |
Does Recursive Knowledge Construction Help Agents Verify Consensus Claims in Science?
arXiv |
2026-10 |
— |
— |
Accept Changes, Reject Attacks: Robust Memory Management for LLM Agents
arXiv |
2026-10 |
— |
— |
"Shall We Approve?" Consequence-Aware Approval for Autonomous Coding Agents
arXiv |
2026-10 |
— |
— |
PrivacyFlow: Privacy in AI Agents through Information Flow Analysis and Consequence-Aware Risk Management
arXiv |
2026-10 |
— |
— |
Pre-Crime for Agents, at the Last Second: Probing When a Recruited Agent Gives In
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
VeriCUA: Verification as Agentic Exploration for Computer-Using Agents
arXiv |
2026-10 |
0 |
— |
Advancing Clinical Genomic Analysis with a Skill-Centred Agent
arXiv |
2026-10 |
0 |
— |
Retrieval as Traversal: Value-Indexed Memory for Stateful Long-Context Agents
arXiv |
2026-10 |
0 |
— |
FIRME: Registry-Anchored Memory Graphs for Evolving, Heterogeneous Investor Knowledge
arXiv |
2026-10 |
0 |
— |
EpiAgent: Epistemic Self-Play for LLM Agents
arXiv |
2026-10 |
0 |
— |
Locating and Closing Information Gaps in Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
Agentic process evidence
arXiv |
2026-10 |
0 |
— |
EvoTeam: Multi-Agent Collaboration via Evolving Teams
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
Long-Term Agent Memory through Complementary Curation and Consolidation
arXiv |
2026-10 |
0 |
— |
Before the Next Rollout: Diagnosing Skill Utility Shifts from Policy Updates
arXiv |
2026-10 |
0 |
— |
What Does "Now" Mean to Your Computer-Use Agent? Stale Screenshots, Page Loads, and Which Check to Run Before a Click
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
MemSteer: Turning User Histories into Supervision for Agent Memory Optimization
arXiv |
2026-10 |
0 |
— |
Goal-Aware Dynamic Multi-Agent Collaboration for Long-Horizon Tasks
arXiv |
2026-10 |
0 |
— |
From extreme weather to agricultural consequences with agentic AI
arXiv |
2026-10 |
0 |
— |
When to Assist and What to Suggest: Joint Gating and Memory-Grounded Hint Generation in One Compact Decoder
arXiv |
2026-10 |
0 |
— |
Gym-Anything for Robotics: Scaling physical simulations with digital agents
arXiv |
2026-10 |
0 |
Project |
The Athlete as a Scientist of One: Agentic Bayesian Experimental Design for Individualized Physiology from Consumer Wearables
arXiv |
2026-10 |
0 |
— |
TaskFrame: Connecting Task-Level Planning with Frame-Level Control for Game Agents
arXiv |
2026-10 |
0 |
— |
Learning Calibrated Confidence for Multi-Agent Collaboration
arXiv |
2026-10 |
0 |
— |
Practical Assessment of Agentic Systems in Acoustics
arXiv |
2026-10 |
0 |
— |
Can AI Agents Interface Clinicians and Complex Models?
arXiv |
2026-10 |
0 |
— |
LLM Agent-Based Design of RISC-V CPUs with FPGA-in-the-Loop Evaluation
arXiv |
2026-10 |
0 |
GitHub · ★ 5 |
Judging Before Propagation: Task-Graph Contextual Integrity for Multi-Agent Privacy
arXiv |
2026-10 |
0 |
— |
Soei: Structured Observability of Evidence and Intent for Multi-Agent AI Software Engineering
arXiv |
2026-10 |
0 |
— |
OPERA: OPamp Design by an Expert-Taught Reasoning Agent, from Specifications to Layout
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Learning to Formulate: Agentic Hardware-Software Co-Design of AI Infrastructure through Progressive Optimization and Verification
arXiv |
2026-10 |
0 |
— |
EmbodiExp: Compositional Experience Transfer for Embodied Agents
arXiv |
2026-10 |
0 |
— |
REALM: Identifying Minimal Task Assets for LLM Agents
arXiv |
2026-10 |
0 |
— |
PCM: A Proof-Carrying Message Protocol for Verifiable Multi-Agent Communication
arXiv |
2026-10 |
0 |
GitHub · ★ 37 |
HERA: Assessing an Autonomous Agent for Quantum Device Measurement
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Don't Lose the Thread: Empowering Long-Horizon LLM Agents with Cognitive Resource Self-Allocation
arXiv |
2026-10 |
0 |
— |
Agentic LLM reasoning in pediatric diabetes outcome prediction from multimodal medical records
arXiv |
2026-10 |
0 |
— |
AgentHeal: Evaluating Goal-Preserving Resource Recovery of Web Agents under Access Barriers
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Aligning Autoresearch: Co-Improvement between Agents and Humans
arXiv |
2026-10 |
0 |
— |
Beyond fabrication: Misattributed numbers in papers written by AI research agents
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Masked Erosion: How Acceptance Gates Lose Redundant Safeguards in Self-Improving Agents
arXiv |
2026-10 |
0 |
— |
SkillCT: A Multidimensional Benchmark and Execution-Free Baseline for Agent Skill Assessment
arXiv |
2026-10 |
0 |
— |
MAS-Loop: Beyond One-Shot Orchestration towards Self-Evolving Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
MUSE: Multi-Level Memory with Selective Runtime Injection for Scientific Coding Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Implicit Minor Detection from Multi-Turn Dialogue in High-Risk AI Interactions: A Constrained Self-Evolving LLM Harness
arXiv |
2026-10 |
0 |
— |
Efficient Agentic Audio-Visual Question Answering through Toolset Pruning and Modality Consolidation
arXiv |
2026-10 |
0 |
— |
Auditable Belief Updates as a Missing Safety Layer for LLM-Powered Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
Think Before You Act: Decision Gate for LLM Agents
arXiv |
2026-10 |
0 |
— |
CALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents
arXiv |
2026-10 |
0 |
— |
Predict the Reward, Not the World: World Reward Models for ML Agents
arXiv |
2026-10 |
0 |
— |
Agent Transfer Protocol: Server-Mediated Agent-to-Agent Communication Architecture for the Agentic Web
arXiv |
2026-10 |
0 |
— |
FACTS-MAS: A Multi-Agent Architecture for Adaptive Forecasting from Heterogeneous Evidence
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Autonomous Quantum Measurement with Agentic AI
arXiv |
2026-10 |
0 |
— |
Tracekit Box: Running AI Agents in Micro-VMs Where Every Way Out Is Decided and Signed
arXiv |
2026-10 |
0 |
GitHub · ★ 42 |
The Code Behind the Layout: AI Agents for Analog Layout Generators
arXiv |
2026-10 |
0 |
— |
WeVisCut: The Whole Editing Room in One Agent From Narration to Long-Form Video
arXiv |
2026-10 |
0 |
— |
A Case Study of a Multi-Agent Kernel-Optimization Loop That Learns a New AI Accelerator
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
What Should a Navigator Remember? Spatiotemporal Memory for Zero-Shot Vision-Language Navigation
arXiv |
2026-10 |
0 |
— |
Skeleton Coding: Program Structure as a Specification for Agents
arXiv |
2026-10 |
0 |
— |
Is it my preference? Role Grounding Attack against Agent's Memory via Social Interaction
arXiv |
2026-10 |
0 |
— |
HEDGE: Gated Hypergraph Memory for Scalable Multi-Agent Debate
arXiv |
2026-10 |
0 |
— |
CoSTA: Co-Evolving Self-Play Timeline Agents for Complex Temporal Knowledge Graph Question Answering
arXiv |
2026-10 |
0 |
— |
Two Birds with One Stone: Skill-Augmented Generation via Unified Retrieval and Utilization
arXiv |
2026-10 |
0 |
— |
Repairs for Whom? Evaluator-Executor Mismatch in Selecting Repairs for Failed LLM Agents
arXiv |
2026-10 |
0 |
— |
GARDEN: Generalist Autonomous Reward Design Through Evidence Navigation for Cooperative Multi-Agent Reinforcement Learning
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
VerifiedLab: Harnessing Knowns for Unknowns
arXiv |
2026-10 |
0 |
— |
Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution
arXiv |
2026-10 |
0 |
— |
RubricAtlas: Mining Type-Level Rubrics from Expert Reports to Evaluate Deep Research Agents
arXiv |
2026-10 |
0 |
— |
When the Critic Falls Behind: Co-Evolving Rubrics and Harnesses for Continual Agent Self-Improvement
arXiv |
2026-10 |
0 |
— |
NextEval: An Agent-Based Solver and Benchmark for Derivative-Free Optimization
arXiv |
2026-10 |
0 |
— |
GEPA-Curator: Curated Learning for Self Improving Agents
arXiv |
2026-10 |
0 |
— |
A Harness Reducing over-Defensive Writing in Agents by Isolating Objections
arXiv |
2026-10 |
0 |
— |
Benchmarking Task Takeover Across Agent Handoffs
arXiv |
2026-10 |
0 |
— |
Your Policy Is Secretly a Harness: The Decision Geometry of Agent Interfaces
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
SkillMem: Compact External Skill Memory for Latent Multi-Agent Reasoning
arXiv |
2026-10 |
0 |
— |
FilingLoop: Evaluating and Evolving LLM Agents for Corporate Disclosure Review and Revision
arXiv |
2026-10 |
0 |
— |
Evaluating and improving AI agents for sequencing protocol understanding
arXiv |
2026-10 |
0 |
— |
Detecting Covert Collusion between AI Agents through Partner-Outsider Positive Listening
arXiv |
2026-10 |
0 |
— |
Measuring Where Intent Gets Lost in Agentic Smart Contract Synthesis
arXiv |
2026-10 |
0 |
— |
What Does an Ontology Catch? Curated, LLM-Generated and Data-Derived Ontologies as Checkers of LLM Agents over Knowledge Graphs
arXiv |
2026-10 |
0 |
— |
WebAnything: From the Wild Web to Scalable Agent Environments
arXiv |
2026-10 |
0 |
GitHub · ★ 6 |
Scan Before You Install: Detecting and Mitigating Local and Composite Vulnerabilities in Skills
arXiv |
2026-10 |
0 |
— |
Poisoned Skills: Propagating Jailbreaks Through Agent Markdown Files
arXiv |
2026-10 |
0 |
— |
When Skills Tell Agents Not to Ask: A Large-Scale Study of Oversight Directives in Agent Skills
arXiv |
2026-10 |
0 |
— |
Rare, Relevant, and Still Wrong: How Memory Fails LLM Group Decisions
arXiv |
2026-10 |
0 |
— |
From Interaction to Capability: Task Advancement as State and Supervision for GUI Agents
arXiv |
2026-10 |
0 |
— |
Concordat: Typed Sessions for LLM Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
M4Bench: Evaluating Procedural Specification for Clinical EHR Derivation Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 49 |
Agent-Conditioned Routing: Diverse Reasoning Paths Within One Frozen Model
arXiv |
2026-10 |
0 |
— |
Grounding LLM Agents in Business Processes: Process-Aware Knowledge Graphs for Reliable Multi-Step Service Invocation
arXiv |
2026-10 |
0 |
— |
Does Your Generalist Skill Optimizer Really Incentivize Agentic Tasks? Rethinking Skill Optimization via Bipartite Attribution
arXiv |
2026-10 |
0 |
GitHub · ★ 4 |
Proactive Memory for Event-Driven LLM Agents
arXiv |
2026-10 |
0 |
Project |
Evolving Agent Teams
arXiv |
2026-10 |
0 |
GitHub · ★ 61 |
Beyond the Evidence Coherence Trap: Agent-Driven Evidence Selection for Temporally Disordered Fake News Detection
arXiv |
2026-10 |
0 |
— |
The View from Nowhere: Population-Centered Agent Evolution
arXiv |
2026-10 |
0 |
— |
Remembering Is Not Enough: Refining Memories and Learning When to Use Them in Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
God Does Not Play Dice: Enforcing Batch Invariance in Agentic Kernel Generation
arXiv |
2026-10 |
0 |
— |
CASEL: Consequence-Aware Selective Execution for Reliable Long-Horizon Tool-Using Agents
arXiv |
2026-10 |
0 |
— |
Monitor Accuracy Isn't the Answer: A Decision-Value Framework for Evaluating AI Agent Restart Monitors
arXiv |
2026-10 |
0 |
— |
When to Intervene? An Empirical Study of Errors and Self-Recovery in LLM Agents
arXiv |
2026-10 |
0 |
Project |
RouteDRAG: Learning to Route and Control Agentic Execution for Multimodal Document RAG
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
Gym-Anything-RT: Scaling Real-Time Environments for Vision-Language Agents
arXiv |
2026-10 |
0 |
— |
Test-Time Communication in Trading Agent Swarms
arXiv |
2026-10 |
0 |
— |
AnalogScout: Simulation-Efficient PVT-Aware Analog Sizing with LLM Agents
arXiv |
2026-10 |
0 |
— |
Where Should the Evaluation Budget Go in LLM-Driven Agent Scaffold Evolution?
arXiv |
2026-10 |
0 |
— |
Spine: Compiling Subtask Trees into Structured Context for Long-Horizon Coding Agents
arXiv |
2026-10 |
0 |
— |
Blindfold: Keeping the Audit Coin Secret from LLM Agents
arXiv |
2026-10 |
0 |
— |
Beyond QASecClaw: Agent Scaling, Verifier Feedback, and Evidence Reliability in Software Assurance
arXiv |
2026-10 |
0 |
— |
From Voiceprints to User Entities: Online Speaker-Memory Construction for Multi-User Spoken Agents
arXiv |
2026-10 |
0 |
— |
Metacognitive Decoding: Efficient Inference for Diffusion Language Agents
arXiv |
2026-10 |
0 |
— |
ETAMem: Decoupling and Aligning Heterogeneous Memory for Long-Term Conversational Agents
arXiv |
2026-10 |
0 |
— |
TrawMem: Reframing Agent Memory as Task-Oriented Role-Adaptive Workspaces
arXiv |
2026-10 |
0 |
— |
DocSkill: Skill-Integrated Agentic Reasoning for Multilingual Multi-Hop Document Question Answering
arXiv |
2026-10 |
0 |
— |
The Trinity: Capability-Driven Recursive Self-Improvement for Multimodal Agents
arXiv |
2026-10 |
0 |
— |
Context Is a Pipeline, Not a Window: Information Attrition in LLM Agent Harnesses
arXiv |
2026-10 |
0 |
— |
Free to Look, Free to Reason: Simple LLM Harnesses Outdiagnose Hand-Crafted Root Cause Analysis Agents
arXiv |
2026-10 |
0 |
— |
Infinity Machine: a self-improving AI agent for research
arXiv |
2026-10 |
0 |
— |
Skill$^{\infty}$: Recursive Self-Improvement for Agent Skill Optimization
arXiv |
2026-10 |
0 |
GitHub · ★ 6 |
RSIV-Harness: A Recursive Self-Improvement Harness Framework for Open-Domain Conversational Text-to-Visualization
arXiv |
2026-10 |
0 |
— |
The Right Fact, the Wrong Decision: SciValidityBench for Scientific Agent Memory
arXiv |
2026-10 |
— |
— |
Harness-Model Collaboration for Long-Horizon Continual Learning
arXiv |
2026-10 |
0 |
— |
Profiling a Frontier Agent with EBench: From Direct Control to Harness with Policy
arXiv |
2026-10 |
0 |
Project |
MarkovSim: an LLM Agent Harness with Bounded Context for Long-Horizon Computational Simulation
arXiv |
2026-10 |
0 |
— |
The Target Agent Is Its Own Best Critic but a Poor Proposer: Self-Improved Red Teaming
arXiv |
2026-10 |
0 |
— |
How Do Coding-Agent Harnesses Self-Evolve? An Empirical Study of Retrospective Adaptation
arXiv |
2026-10 |
0 |
— |
The Physical Cost of Context: Quality-Constrained Resource Accounting for Coding Agents
arXiv |
2026-10 |
0 |
— |
Quantifying Claim Smoothing by LLM Research Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
OA-RAG: Ontology-based Agentic RAG for Faithful LLM Reasoning
arXiv |
2026-10 |
0 |
— |
Ontology-guided multi-agent reasoning for explainable differential diagnosis
arXiv |
2026-10 |
0 |
— |
When Retrieval Is Not Enough: Obsolete Facts in Agent Memory
arXiv |
2026-10 |
0 |
— |
Skill Dreaming: Evolving Repository Skills through Pull Request Streams
arXiv |
2026-10 |
0 |
— |
CohortFlow: A Steerable Clinical Cohort Curation Agent with Localized Semantic Correction
arXiv |
2026-10 |
0 |
— |
Stratum: Repository Memory for Coding Agents Beyond the Context Window
arXiv |
2026-10 |
0 |
— |
Replication or Corroboration? An Axiomatic Framework for Evidential Dependence in Multi-Agent Debate
arXiv |
2026-10 |
0 |
— |
When Connectivity Misleads: Guiding Agentic Search with Contrastive Evidence Paths
arXiv |
2026-10 |
0 |
— |
Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol
arXiv |
2026-10 |
0 |
— |
Do Coding Agents Still Need a Debugger? Frontier-Model-Based Fault Localization at the Accuracy Ceiling
arXiv |
2026-10 |
— |
— |
PolicyMediator: Multi-Regional Policy Learning through Mediation Agents with Evolving Skills in OPC Ecosystems
arXiv |
2026-10 |
0 |
— |
Generation and Verification: Human-Agent Collaboration for Safety-Critical Automotive Software Engineering
arXiv |
2026-10 |
0 |
— |
EvoRouter: Cost-Efficient Multi-Agent Routing via Insight-Guided On-Policy Distillation
arXiv |
2026-10 |
0 |
GitHub · ★ 5 |
Agentic Algorithm Design: LLMs that Propose, Formalise, and Prove Runtime Bounds for Evolutionary Algorithms in Lean
arXiv |
2026-10 |
0 |
— |
Belief-Conditioned Multi-Agent Actor Critic for Decentralized LLM Collaboration
arXiv |
2026-10 |
0 |
— |
Proof-of-Agentness: Towards Verifying Agents on the Web
arXiv |
2026-10 |
0 |
— |
AutoEnergyDesign: An LLM-Driven Agent for Automatic Energy System Design Optimization
arXiv |
2026-10 |
0 |
— |
LongSearchEval: Evaluating Long-Horizon Web Search Capabilities of Large Language Models via a Trajectory-Aware Agent Framework with Failure-Node Localization
arXiv |
2026-10 |
0 |
— |
Collective Emergent Alignment: Measuring and Amplifying Interaction-Induced Reasoning Alignment in Multi-Agent LLMs
arXiv |
2026-10 |
0 |
— |
Scope-Gap Injection: Exploiting Evaluator Channel Blind Spots in the OpenClaw Agent Security Competition
arXiv |
2026-10 |
0 |
— |
SKALOS: Learning to Distinguish Similar Skills with Capability Graphs
arXiv |
2026-10 |
0 |
— |
CEPO: A General Framework for Improving Prompt Optimization in Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
KAPPA: Checking the Proofs in Statistics Papers with Lean 4 and Language Model Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 11 |
The Hidden Costs of Forgetting: A Study of Trajectory Deletion in Software Engineering Agents
arXiv |
2026-10 |
0 |
— |
Autonomous Structural Discovery Of Bioprocess Dynamics And Kinetics Via LLM-Driven Agents
arXiv |
2026-10 |
0 |
— |
RepoTrace: Improving Requirement Tracing in Multi-Round Agent Interactions
arXiv |
2026-10 |
— |
— |
Joint Latent Reasoning and Observation for Agentic Video Understanding
arXiv |
2026-10 |
0 |
— |
CoSA: The Conflict of Skills Attack in Real-World LLM Agent Ecosystems
arXiv |
2026-10 |
0 |
— |
AskGLP1: A Knowledge Graph LLM Agent for Exploring GLP-1 Experience from Online Patient Narratives
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
GUIFlow: Temporal Redundancy-Aware Incremental Inference for GUI Agents
arXiv |
2026-10 |
0 |
— |
AIPC: Evidence-Guided Agent Workflows for PyTorch Project Deployment
arXiv |
2026-10 |
0 |
— |
Adaptive Retrieval, Selective Answers: Evaluating Conformal Factuality in Agentic RAG
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Silent and Costly: Characterizing and Detecting Prompt-Caching Mis-Integration in LLM Agents
arXiv |
2026-10 |
0 |
— |
SceneAgent: An Agentic Pipeline from 3D and 4D Captures to Simulation Environments
arXiv |
2026-10 |
0 |
— |
Agent Memory as Recommendation: Recommender-System Design for Long-Term Memory of LLM Agents
arXiv |
2026-10 |
0 |
— |
Thinking-in-Bets at Test Time Reduces Sycophantic Revisions in LLM Agents
arXiv |
2026-10 |
0 |
Project |
Agentic Systems for Calibrating Cosmological Simulations
arXiv |
2026-10 |
— |
GitHub · ★ 0 |
OmniBrain-0: A Unified Agentic Model for Embodied Intelligence
arXiv |
2026-10 |
0 |
Project |
Hierarchical Skill Evolution for Coding Agent
arXiv |
2026-10 |
0 |
— |
From Questions to Queries to Insights: A Schema-Grounded Agentic Workflow for Natural-Language Crash Data Analysis
arXiv |
2026-10 |
0 |
— |
Semantic Atoms for Scalable Text Analysis in Data Analytics in the Era of LLM-Based Agents
arXiv |
2026-10 |
0 |
— |
BehaviorMem: Building and Evaluating User Memory from Real-World E-Commerce Behavior Streams
arXiv |
2026-10 |
0 |
— |
Rethinking Verification in Agent Harnesses
arXiv |
2026-10 |
0 |
— |
MatOS: An Agent Operating System for Materials Research
arXiv |
2026-10 |
0 |
GitHub · ★ 12 |
AgentProofBench: Benchmarking Reusable Verifiers for Web Agents
arXiv |
2026-10 |
0 |
— |
SentryBench: A Cost-Sensitive Benchmark for Adversarial Robustness of Tool-Using LLM Agents in Public-Safety Deployments
arXiv |
2026-10 |
0 |
— |
EvolveMAS: A Unified Self-Evolution Framework for Decentralized Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
The Coordination Tax and the Heterogeneity Dividend: Break-Even Conditions for Multi-Agent LLM Pipelines
arXiv |
2026-10 |
0 |
— |
EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis
arXiv |
2026-10 |
0 |
— |
AgentBaton: Efficient Model Collaboration in Agentic Tasks
arXiv |
2026-10 |
0 |
— |
Agentic AI Orchestration Framework for Spatial Intelligence: Supply Chain Deployment
arXiv |
2026-10 |
0 |
— |
CaliMem: Search-Efficient Data Agent via Calibrated Memory
arXiv |
2026-10 |
0 |
— |
What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents
arXiv |
2026-10 |
0 |
— |
CTXAgent - Self-Controlled Context Management of LLM Agents
arXiv |
2026-10 |
0 |
— |
Matched Capabilities, Different Behaviors: A Controlled Study of MCP and CLI Interfaces for Language Agents
arXiv |
2026-10 |
0 |
— |
MultiEvoSkill: Structure-Aware Multi-Agent Skill Evolution for Generalizable Skill Artifacts
arXiv |
2026-10 |
0 |
— |
CodeMeepo: A Multi-Agent Workflow for Repository-Level Code Migration and Optimization
arXiv |
2026-10 |
0 |
— |
LLM WikiLeaks: Uncovering URL-Based Capability Extension and Security Circumvention in Agentic Message Boards
arXiv |
2026-10 |
0 |
— |
The Price of Scale in Proactive Memory Recall
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
MulFCoder: Framework-conditioned Multi-agent for MLLM-based Multi-framework Front-end Code Generation
arXiv |
2026-10 |
0 |
— |
ParetoRSI-Gym: Can AI Agents Improve Language Models Across Multiple Domains?
arXiv |
2026-10 |
0 |
— |
Outcome-Aware Agents: When Does Causal Structure Help a Forward Model Foresee the Consequences of Actions?
arXiv |
2026-10 |
— |
GitHub · ★ 0 |
AndroidFlux: Evaluating and Reward Modeling Failure Recovery Capabilities in Mobile Use Multimodal Agents
arXiv |
2026-10 |
0 |
— |
SEEK: Skill Extraction with Evolving Knowledge against Frontier Science Agents
arXiv |
2026-10 |
0 |
— |
NameTrap: Steering AI Coding Agents Toward Security Vulnerabilities via Variable Names
arXiv |
2026-10 |
0 |
— |
$\Delta\tau$-Bench: Pushing Agents Toward a Moving Capability
arXiv |
2026-10 |
0 |
— |
Holding a Steady Course: Knowledge Evolution from Expertise and Lessons for Continual Agent Learning
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Memory Tax: Dissecting the Net Value of Agent Memory
arXiv |
2026-10 |
0 |
Project |
How Do Small Coding Agents Use Code Graphs and Embeddings?
arXiv |
2026-10 |
0 |
— |
Build, Don't Play: Self-Improving LLM Research Agents Develop a Hearthstone Battlegrounds AI That Rivals Top Players
arXiv |
2026-10 |
0 |
GitHub · ★ 228 |
PACE: Research Proving Agent for Adaptive Conjecture Exploration
arXiv |
2026-10 |
0 |
— |
SkillShift-Bench: Benchmarking Skill Learning under Environment Evolution for Web Agents
arXiv |
2026-10 |
0 |
Project |
LegoSR: A Unified Framework for Self-improving Agent-Guided Scientific Discovery
arXiv |
2026-10 |
0 |
GitHub · ★ 33 |
Build or Adopt? Custom-Built vs. General-Purpose LLM Agent Harnesses for Automotive GUI Testing
arXiv |
2026-10 |
0 |
— |
RSI Arena: Towards Recursive Self-Improvement with Human Feedback
arXiv |
2026-10 |
0 |
Project |
OS-Wizard: Scaling Computer-Using Agents to Specialized Software via Knowledge-Guided Exploration
arXiv |
2026-10 |
0 |
— |
ReGE: Stronger Agents through Experience Synthesis
arXiv |
2026-10 |
0 |
— |
Learning When to Collaborate: Selective Multi-Agent Medical Reasoning via Uncertainty-Aware Routing
arXiv |
2026-10 |
0 |
— |
CoEvolve: Counterfactual-Guided Co-Evolution of Memory and Collaboration in Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
FraudSociety: A Generative Agent-Based Society for the Financial-Fraud Economy
arXiv |
2026-10 |
0 |
— |
ConMem: Counterfactual Multi-Scenario Probing for Resolving LLM Memory Misalignment
arXiv |
2026-10 |
0 |
— |
AgentOrch: How Close Can Global Orchestration Come to Step-by-Step Interaction?
arXiv |
2026-10 |
0 |
— |
FlashScript: Compiling Recurring Agent Workflows into Reusable Neuro-Symbolic Programs
arXiv |
2026-10 |
0 |
— |
Strategic Calibration in LLM Multi-Agent Systems: Audited Pivot Scoring for Truthful Elicitation
arXiv |
2026-10 |
0 |
— |
ProbeID: Verifying MCP Tool Identity for LLM Agents under Registry Drift
arXiv |
2026-10 |
0 |
— |
Can Agentic Code Review Catch Vulnerabilities Before They Are Merged?
arXiv |
2026-10 |
— |
GitHub · ★ 4 |
Novel or Not? PatEx-Gym: Agentic Prior-Art Search for Novelty Assessment in Patent Examination
arXiv |
2026-10 |
0 |
— |
Distill, Debate, Discover: Grounding Multi-Agent Research Ideation in Graph-Distilled Insights
arXiv |
2026-10 |
0 |
— |
Estimation of Propagated Risk in LLM-based Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
AdaPRA: Process Reward Agents with Adaptive Rubrics for Data Science
arXiv |
2026-10 |
0 |
— |
A Deterministic Security Runtime for LLM Agents on MPI and Batch-Scheduled HPC Systems: Consensus-Governed, Hardware-Rooted Supervision of Every Rank
arXiv |
2026-10 |
— |
— |
Responsibility Boundaries for Agentic Systems: A Comparative Evaluation of Probabilistic Agentic Programming
arXiv |
2026-10 |
0 |
— |
Operationalizing Governed Agentic Workflows in Regulated Finance: A Customer Zero Case Study using Zafin AIOS
arXiv |
2026-10 |
0 |
— |
Cross-Platform Application Migration to HarmonyOS: An Agentic Pipeline with Artifact-Derived Verification
arXiv |
2026-10 |
0 |
— |
Multi-Agent Retrieval-Augmented Causality Assessment (MARCA) of Individual Case Safety Reports in Pharmacovigilance
arXiv |
2026-10 |
0 |
— |
FlexA-Mem: Budget-Adaptive Agentic Memory for LLM Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Certified Execution Harness for Agentic AI: Safe and Least-Intrusive Petri-Net Supervisory Control Between Reasoning and Action
arXiv |
2026-10 |
0 |
— |
RIDAR: Safeguarding Research Integrity in Autonomous Research Agents via Automated Detection and Runtime Intervention
arXiv |
2026-10 |
0 |
— |
Harnessing Language Models as Optimization Modeling Agents
arXiv |
2026-10 |
0 |
— |
ADAPT for Science: A Multi-Agent System for Self-Evolving Scientific Optimization
arXiv |
2026-10 |
0 |
— |
Governed Self-Evolution for Autonomous Investing: A Multi-Agent LLM System with an Invariant Core
arXiv |
2026-10 |
0 |
GitHub · ★ 14 |
HERD: Harness-Evolving Recursive Distillation for Continual Learning
arXiv |
2026-10 |
0 |
GitHub · ★ 4 |
From Prose to Properties: Testing Research Claims with ARATest for Agent-Native Science
arXiv |
2026-10 |
0 |
GitHub · ★ 6 |
RoleFlow: Self-Evolving Multi-Role Agentic Collaboration via Structural Adaptation
arXiv |
2026-10 |
0 |
— |
GAV: A scientific experimentation environment for PDE agents
arXiv |
2026-10 |
0 |
GitHub · ★ 6 |
HGM-Dual: Lineage-Aware Guided Agent Harness Optimization for Long-Horizon Planning
arXiv |
2026-10 |
0 |
— |
GuideEvolve: Self-Evolving Agent Framework with Expert Guidance and 3Ts Mechanisms
arXiv |
2026-10 |
0 |
— |
Real PMIDs, Wrong Papers: Harness Design for Biomedical Claim-Verification Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 3 |
PreCoMem: Predictive Cognitive Memory for Self-Evolving Long-Term Dialogue Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Recursive Self-Improvement on a Budget: Scoring Code Edits Before They Run
arXiv |
2026-10 |
0 |
— |
Harness Scaling for Generalizable Agentic Reinforcement Learning
arXiv |
2026-10 |
0 |
— |
TrainBridge: Agentic NPU Training Optimization Guided by GPU Compilation Evidence
arXiv |
2026-10 |
0 |
— |
Evolving with Open Worlds: Towards Recursive Self-Improvement for Future Forecasting
arXiv |
2026-10 |
0 |
— |
BIRD-RL: Scaling Agentic Reinforcement Learning over Stateful Data-Centric Environments
arXiv |
2026-10 |
0 |
— |
Harness Self-Evolution for Long-Horizon Terminal Agents
arXiv |
2026-10 |
0 |
— |
MaD-RSI: Model-as-Data for Recursive Self-Improvement of Agent Harnesses
arXiv |
2026-10 |
0 |
— |
COMPASS: Routing via Capability-Oriented Model Profiling with Adaptive Search, Selection, and Tool Use
arXiv |
2026-10 |
0 |
— |
Evolving Agent Interfaces
arXiv |
2026-10 |
0 |
GitHub · ★ 12,861 |
Self-Evolving Static Analysis via LLM Pseudo-Labeling
arXiv |
2026-10 |
0 |
— |
AgentPolicyForge: From Executable Environments to Policy Benchmarks and Adaptive Guards
arXiv |
2026-10 |
0 |
— |
Self-Improving Agents with World Model Predictive Paths
arXiv |
2026-10 |
0 |
— |
Forging More Tasks into Better Harnesses: Agentic Harness Evolution via Successive Scale Racing
arXiv |
2026-10 |
0 |
— |
FALCA: Falsification-Aware Reinforcement Learning with Hierarchical Credit Assignment for Long-Horizon Deep Search Agents
arXiv |
2026-10 |
0 |
— |
ChartEvo: Self-Evolving Agent Harnesses for Long-Tail Chart Reasoning
arXiv |
2026-10 |
0 |
— |
DuplexLedger: Benchmarking Transactional Memory in Full-Duplex Speech Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 4 |
RaspAssistant as an Agentic Framework for Automating Adsorption Simulation Research in Porous Materials
arXiv |
2026-10 |
0 |
— |
Recoverability and Verifier Strength: Two Measured Preconditions for Unattended Agent Self-Improvement
arXiv |
2026-10 |
0 |
— |
KPHarness: Co-Evolving Knowledge and Probing for Long-Horizon Gaming Harnesses
arXiv |
2026-10 |
0 |
— |
GBClaw: An agentic AI framework for atomistic grain boundary research and mechanistic discovery
arXiv |
2026-10 |
0 |
— |
VisualEvolve: A Skill Self-Evolution Framework for Visual Reasoning
arXiv |
2026-10 |
0 |
— |
World2HOSI: A Multi-Expert Harness for Long-Horizon Human-Object-Scene Interaction Generation
arXiv |
2026-10 |
0 |
— |
Self-Evolving Agents through Privileged Retrospective Distillation
arXiv |
2026-10 |
0 |
— |
Measure First, Then Allocate: Calibrated Rollout Budgets for MCP Agents in Reinforcement Learning
arXiv |
2026-10 |
0 |
— |
Object-Oriented Harness: Evolving Agent Harnesses through Object-Oriented Design
arXiv |
2026-10 |
0 |
— |
Learning What to Remember: Resource-Efficient Self-Improvement for Lifelong Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 3 |
Reliable Before Profitable: QuantHarness for Governing Financial Strategy Development with LLM Agents
arXiv |
2026-10 |
0 |
— |
EvoSkill v2: Fully Agentic Skill Evolution from Benchmark Adaptation to a Stateful Evolver
arXiv |
2026-10 |
0 |
— |
OpenSaC: A Python Library for Search as Code in Deep Research Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 4 |
OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory
arXiv |
2026-10 |
0 |
— |
HarnessRL: Co-Evolving Policies with a Just-in-Time Training Harness for Open-Ended RL
arXiv |
2026-10 |
0 |
GitHub · ★ 61 |
HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs
arXiv |
2026-10 |
0 |
GitHub · ★ 142 |
Cassette-Editor: An Agent Harness for Editing Video the Way You Do
arXiv |
2026-10 |
— |
— |
Epi: Towards a Persistent, Self-Improving Agent for Autonomous Research
arXiv |
2026-10 |
0 |
GitHub · ★ 12 |
Design and Cross-Domain Evaluation of an Agent Execution Harness
arXiv |
2026-10 |
0 |
— |
Symbolic Constraint Agents: Mining Formal Tool-Call Constraints to Prevent Agent Errors
arXiv |
2026-10 |
0 |
— |
Bud ContextCompact 2.0: Context Compaction for Agent Success and Accuracy with Small Language Models and Constrained Context Windows
arXiv |
2026-10 |
0 |
— |
Bridging Concepts: Multi-Agent Reasoning over Scientific Concept Graphs for High-Impact Research Prediction
arXiv |
2026-10 |
0 |
— |
Compile Once, Run Cheap: Frozen, Portable Skills for Embodied Agents, and What Recursive Self-Improvement Adds
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
Trackable Skill Library Lifecycle Management for Agentic Reinforcement Learning
arXiv |
2026-10 |
0 |
— |
CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow?
arXiv |
2026-10 |
0 |
— |
Who Convinced Whom? Tracing How LLMs Change Their Minds in Multi-Agent Systems
arXiv |
2026-10 |
0 |
— |
ReTRACE: Feedback-Guided LLM Agent in a Chemistry Harness for Retrosynthesis Planning
arXiv |
2026-10 |
0 |
— |
Selm: Self-Evolving Library Management for Multi-Skill Agent Optimization
arXiv |
2026-10 |
0 |
— |
Learning to Compose: Recursive Self-Improvement of Shared Executable Stacks for Zero-Shot Long-Horizon Robot Manipulation
arXiv |
2026-10 |
0 |
— |
MELD: Recursive Self-Improvement via Hierarchical Coordination for Multi-Agent Scientific Discovery
arXiv |
2026-10 |
0 |
— |
Hippo: Training-Free Agent Self-Evolution via Hippocampal Replay
arXiv |
2026-10 |
0 |
— |
Organon : From Evolutionary Search to Recursive Self-Improving Agents for Scientific Discovery
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Self-Evolving Coding Agents via Harness-Policy Optimization
arXiv |
2026-10 |
0 |
GitHub · ★ 56 |
DrugHarness: An Evolvable Agent Harness for Computational Drug Discovery
arXiv |
2026-10 |
0 |
— |
From Trajectories to Differentiated Skills: Cognitive Differentiation for Self-Improving Agents
arXiv |
2026-10 |
0 |
— |
Darwin-Claw: Environment-Adaptive Experience Consolidation for Recursively Self-Evolving Agents
arXiv |
2026-10 |
0 |
— |
Optimist Deep Research: An Iterative, State-Centric Agent for Autonomous Literature Research
arXiv |
2026-10 |
0 |
— |
WebCoder: Scaling General Frontend Development with Artifact-Grounded Agentic Reinforcement Learning
arXiv |
2026-10 |
0 |
— |
Reward Engine: Reward-Centric Self-Evolution for Open-Ended Tool Agents
arXiv |
2026-10 |
0 |
— |
MuninAI: A Self-Hosted Agentic Framework for Independent Research Groups
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
CRAFT: Corrective and Robust Multi-Agent Framework for Text-to-Parametric CAD
arXiv |
2026-10 |
— |
GitHub · ★ 1 |
Federation over Harness
arXiv |
2026-10 |
0 |
— |
Agentic Interpretation: Harness-Governed Evidence Construction for Software Agents
arXiv |
2026-10 |
0 |
— |
R$^2$Orch: Recursive Self-Improvement for Agent Orchestration via Self-Play within a Skill-Internalization Loop
arXiv |
2026-10 |
0 |
GitHub · ★ 13 |
NoveltyScope: A Multi-Agent Research Assistant for Evidence-Grounded Novelty Assessment of Scientific Papers
arXiv |
2026-10 |
0 |
— |
Skip the Compressor: Coding Agents Fix More Issues When They Pick Whole Files
arXiv |
2026-10 |
0 |
— |
A Self-Improving Agent that Grows its Own DSL: Oracle-Free Operator Induction for Exact Question Answering
arXiv |
2026-10 |
0 |
— |
CausalEvo: Causal Skill Self-Evolution for Health LLM Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Verifier-Harness Co-Evolution for Training-Free Self-Improvement on Open-Ended Tasks
arXiv |
2026-10 |
0 |
GitHub · ★ 1 |
SEAR: Self-Evolving Agentic Robotics
arXiv |
2026-10 |
0 |
— |
Coupled Harness--Model Evolution for Scientific Equation Discovery
arXiv |
2026-10 |
0 |
— |
Memorix: Managing Memory Selections for AI Agents
arXiv |
2026-10 |
— |
— |
Drafting Agents, Not Just Tokens: Speculative Recursive Self-Improvement for Agentic Systems
arXiv |
2026-10 |
0 |
GitHub · ★ 6 |
Lessons from Luck: Do LLM-Written Lessons Carry Chance into an Agent's Next Decision?
arXiv |
2026-10 |
0 |
— |
Exposure Is Not Commitment: Double-Stratified Evaluation of Latent Monitors for Tool-Using Agents
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Harmony Harness: Multi-Objective Self-Evolution of Agent Harnesses
arXiv |
2026-10 |
0 |
— |
Triage-PiEvo: Evidence Admission for Principle-Evolving Scientific Agents
arXiv |
2026-10 |
0 |
— |
Evolving to Match Capabilities: Improving Harness Self-Evolution via Skill Adaptation
arXiv |
2026-10 |
0 |
Project |
Proactive Agentic Orchestration
arXiv |
2026-10 |
0 |
GitHub · ★ 2 |
Harness^2: Recursive Agent Harnessing for an Open World
arXiv |
2026-10 |
1 |
GitHub · ★ 22,232 |
RSI-Anything: Unlocking Recursive Self-Improvement with World Models
arXiv |
2026-10 |
0 |
GitHub · ★ 1,106 |
Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks
arXiv |
2026-10 |
0 |
— |
Harness Evolution Hits a Ceiling: When Weight Training Should Begin
arXiv |
2026-10 |
0 |
— |
Recursive Self-Improvement through Multi-Agent Self-Supervision
arXiv |
2026-10 |
0 |
— |
ReSI: Recursive Safety Improvement toward Resistant and Resilient AI
arXiv |
2026-10 |
0 |
— |
SLICE-Mem: Salience-Gated Lightweight Information Curation for Efficient LLM Agent Memory
arXiv |
2026-10 |
0 |
— |
UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Recursive Semantic Divergence for LLM Agent Consistency
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
RADAgent: Automating RADseq Pipelines with Private Open-Source LLMs
arXiv |
2026-10 |
0 |
GitHub · ★ 0 |
Stateless Language Agents: Scaling Long-Horizon Automated Research
arXiv |
2026-10 |
0 |
— |
ImproveAnyTask: An Autonomous Post-Training Harness for Iterative Model Self-Improvement
arXiv |
2026-10 |
0 |
— |
AgentDiscover: Autonomous Discovery with Minimal Search Scaffolding
arXiv |
2026-10 |
0 |
GitHub · ★ 8 |
Complex Agents, Shallow Tests: Demystifying and Enhancing Test Adequacy of Agent Harness in the Wild
arXiv |
2026-10 |
0 |
— |
MESH-Harness: Self-Improving Agent Harnesses via Bandit-Guided Compositional Evolution
arXiv |
2026-10 |
0 |
— |
Recursive Harness Self-Improvement for Frontier Reasoning Data Synthesis
arXiv |
2026-10 |
0 |
— |
Learning What to Investigate Next: Meta-Reasoning for Long-Horizon Research Agents
arXiv |
2026-10 |
0 |
— |
Safety Must Survive Self-Improvement: Why Failures Persist and How Agents Recover
arXiv |
2026-10 |
0 |
— |
Auditing Web Agent Evaluation on WebArena-Lite: Human Review of Outcomes and Trajectories
arXiv |
2026-10 |
0 |
— |
RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement
arXiv |
2026-09 |
0 |
— |
TIAgent: From Natural Language to Grounded Agentic Workflows for Computational Pathology
arXiv |
2026-09 |
0 |
— |
SelfSearch: Reward-Free Search for Self-Improving Agents
arXiv |
2026-09 |
0 |
— |
Self-Evolving Harness on Multiple Tasks with the Agent as Its Own Optimizer
arXiv |
2026-09 |
0 |
— |
Mixture of Self-Improving Branches For Agent Harness Optimization
arXiv |
2026-09 |
0 |
— |
Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI
arXiv |
2026-09 |
0 |
— |
Collaborative Principle Evolution via Evidence Transfer for Scientific Discovery
arXiv |
2026-09 |
0 |
— |
Recursive Harness Distillation across Agents for Robot Manipulation
arXiv |
2026-09 |
0 |
— |
Breaking the Environment Wall: A Unified Framework for Preparing and Evolving Agent-Native Environments
arXiv |
2026-09 |
0 |
— |
LLM Agents Can Easily Tamper With Their Own Traces
arXiv |
2026-09 |
0 |
— |
Recursive self-improvement of AI research agents
arXiv |
2026-09 |
0 |
— |
RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
arXiv |
2026-09 |
0 |
GitHub · ★ 1,707 |
Total Cost of Agency: Exact Attribution of Memory Injection Cost in Multi-Agent LLM Workflows
arXiv |
2026-09 |
0 |
GitHub · ★ 0 |
Bridging Static and Agentic RAG for Taiwanese Historical Question Answering
arXiv |
2026-09 |
0 |
— |
SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness
arXiv |
2026-09 |
0 |
GitHub · ★ 3,631 |
ModularRSI: Modular and Generalizable Recursive Harness Self-Improvement
arXiv |
2026-09 |
0 |
— |
RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments
arXiv |
2026-09 |
0 |
— |
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
arXiv |
2026-09 |
0 |
GitHub · ★ 1,694 |
Harness-of-Harness: Multi-Day Autonomous Software Development with Continual Improvement
arXiv |
2026-09 |
0 |
GitHub · ★ 190 |
HarnessEvolve: Learning from Reference Trajectories for Reliable Agent Self-Evolution
arXiv |
2026-09 |
0 |
— |
AutoRef: Harness Optimization for Agentic Multi-Reference Image Generation
arXiv |
2026-09 |
0 |
GitHub · ★ 34 |
WHALE: A Simple Recipe for Joint Harness-Weight Optimization
arXiv |
2026-08 |
0 |
GitHub · ★ 63 |
Context as an Environment: Programmatic Context Management for Long-Horizon Agents
arXiv |
2026-08 |
0 |
— |
Agent Lightning v1.0: Towards Harnessed Agentic RL
arXiv |
2026-08 |
0 |
— |
Self-Evolving Embodied Agents via Skill-Harness Evolution
arXiv |
2026-08 |
0 |
— |
PAST-Bench: Benchmarking the Foundations of Recursive Self-Improvement in Personal Agents
arXiv |
2026-08 |
0 |
GitHub · ★ 35 |
HarnessCompass: Guiding Automatic Harness Evolution toward Generalizable and Effective Agent Harnesses
arXiv |
2026-08 |
0 |
— |
PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents
arXiv |
2026-08 |
4 |
— |
Recursive Experiential-Working Memory Evolution for Long-Horizon Agent Harnesses
arXiv |
2026-08 |
10 |
GitHub · ★ 265 |
Prime Agent: A Self-Improving RLM Harness
arXiv |
2026-08 |
17 |
GitHub · ★ 22,554 |
Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops
— |
2026-07 |
38 |
— |
Kimi K3: Open Frontier Intelligence
arXiv |
2026-07 |
103 |
GitHub · ★ 9,041 |
Recursive harness self-improvement
arXiv |
2026-07 |
31 |
— |
ACM: Agentic Context Management for Long Horizon Tasks
arXiv |
2026-07 |
13 |
GitHub · ★ 46 |
CompactionRL: Reinforcement Learning with Context Compaction for Long-Horizon Agents
arXiv |
2026-07 |
15 |
— |
Structured Feedback Improves Repair in an LLM Agent Loop
arXiv |
2026-07 |
3 |
— |
Self-Improvements in Modern Agentic Systems: A Survey
— |
2026-07 |
19 |
GitHub · ★ 604 |
MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution
arXiv |
2026-07 |
23 |
— |
Rethinking the Evaluation of Harness Evolution for Agents
arXiv |
2026-07 |
51 |
GitHub · ★ 50 |
Frontis-MA1: Training an AI4AI Model towards Recursive Self-Improvement in Machine Learning Engineering
arXiv |
2026-07 |
8 |
GitHub · ★ 870 |
Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy
arXiv |
2026-06 |
0 |
GitHub · ★ 54 |
HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
arXiv |
2026-06 |
49 |
— |
The Past Is Prologue: A Plug-in Controller for Selective Updates in Sequentially Evolving LLM Memory
arXiv |
2026-06 |
1 |
— |
From Question Answering to Task Completion: A Survey on Agent System and Harness Design
arXiv |
2026-06 |
13 |
— |
Agent Harness for Large Language Model Agents: A Survey
Preprints |
2026 |
3 |
GitHub · ★ 377 |
Scaffold Effects on GAIA: A Controlled Comparison
arXiv |
2026-06 |
7 |
— |
Context Compression for LLM Agents: A Survey of Methods, Failure Modes, and Evaluation
Preprints |
2026 |
0 |
— |
The Verification Horizon: No Silver Bullet for Coding Agent Rewards
arXiv |
2026-06 |
8 |
— |
Self-Harness: Harnesses That Improve Themselves
arXiv |
2026-06 |
100 |
GitHub · ★ 137 |
Stop Comparing LLM Agents Without Disclosing the Harness
Second Workshop on Agents in the Wild: Safety, Security, and Beyond |
2026 |
8 |
— |
Are We Ready For An Agent-Native Memory System?
arXiv |
2026-06 |
17 |
GitHub · ★ 150 |
Evolve as a Team: Collaborative Self-Evolution for LLM-based Multi-Agent Systems
arXiv |
2026-05 |
0 |
GitHub · ★ 64 |
Learning from Repair Experience: Self-Evolving Agents for Automated Vulnerability Repair
ICSE 2027 |
2026-05 |
0 |
GitHub · ★ 0 |
MOSS: Self-Evolution through Source-Level Rewriting in Autonomous Agent Systems
arXiv |
2026-05 |
12 |
GitHub · ★ 29 |
Ask Early, Ask Late, Ask Right: When Does Clarification Timing Matter for Long-Horizon Agents?
arXiv |
2026-05 |
5 |
— |
From Raw Experience to Skill Consumption: A Systematic Study of Model-Generated Agent Skills
arXiv |
2026-05 |
20 |
— |
Code as Agent Harness
arXiv |
2026-05 |
47 |
GitHub · ★ 768 |
SkillOS: Learning Skill Curation for Self-Evolving Agents
arXiv |
2026-05 |
53 |
— |
Adapting the Interface, Not the Model: Runtime Harness Adaptation for Deterministic LLM Agents
arXiv |
2026-05 |
29 |
GitHub · ★ 227 |
LoopTrap: Termination Poisoning Attacks on LLM Agents
arXiv |
2026-05 |
1 |
— |
Learning Agent-Compatible Context Management for Long-Horizon Tasks
arXiv |
2026-05 |
15 |
— |
SpecBench: Measuring Reward Hacking in Long-Horizon Coding Agents
arXiv |
2026-05 |
29 |
GitHub · ★ 17 |
Toward autonomous long-horizon engineering for ML research
arXiv |
2026-04 |
14 |
GitHub · ★ 146 |
Squeez: Task-Conditioned Tool-Output Pruning for Coding Agents
arXiv preprint arXiv:2604.04979 |
2026-04 |
2 |
GitHub · ★ 24 |
Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems
arXiv |
2026-04 |
48 |
GitHub · ★ 2,210 |
Escher-Loop: Mutual Evolution by Closed-Loop Self-Referential Optimization
arXiv |
2026-04 |
9 |
GitHub · ★ 8 |
Reinforced Agent: Inference-Time Feedback for Tool-Calling Agents
Proceedings of the Fifth Workshop on Generation, Evaluation and Metrics (GEM) |
2026-04 |
7 |
— |
From Agent Loops to Structured Graphs:A Scheduler-Theoretic Framework for LLM Agent Execution
arXiv |
2026-04 |
3 |
— |
ContextBudget: Budget-Aware Context Management for Long-Horizon Search Agents
arXiv preprint arXiv:2604.01664 |
2026-04 |
14 |
GitHub · ★ 10 |
ContextWeaver: Selective and Dependency-Structured Memory Construction for LLM Agents
arXiv preprint arXiv:2604.23069 |
2026-04 |
4 |
— |
Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering
arXiv preprint arXiv:2604.08224 |
2026-04 |
73 |
— |
Harness Engineering for Language Agents: The Harness Layer as Control, Agency, and Runtime
Preprints |
2026-03 |
0 |
— |
Meta-Harness: End-to-End Optimization of Model Harnesses
arXiv |
2026-03 |
305 |
GitHub · ★ 1,248 |
Natural-Language Agent Harnesses
arXiv |
2026-03 |
58 |
— |
Bilevel Autoresearch: Meta-Autoresearching Itself
arXiv |
2026-03 |
8 |
GitHub · ★ 209 |
Schema First Tool APIs for LLM Agents: A Controlled Study of Tool Misuse, Recovery, and Budgeted Performance
arXiv |
2026-03 |
8 |
GitHub · ★ 0 |
DARWIN: Dynamic Agentically Rewriting Self-Improving Network
arXiv |
2026-02 |
1 |
GitHub · ★ 0 |
ReCreate: Reasoning and Creating Domain Agents Driven by Experience
ACL 2026 |
2026-01 |
0 |
GitHub · ★ 169 |
SWE-Pruner: Self-Adaptive Context Pruning for Coding Agents
arXiv preprint arXiv:2601.16746 |
2026-01 |
56 |
GitHub · ★ 319 |
Beyond Static Summarization: Proactive Memory Extraction for LLM Agents
arXiv preprint arXiv:2601.04463 |
2026-01 |
20 |
— |
Memory in the Age of AI Agents
arXiv preprint arXiv:2512.13564 |
2025-12 |
297 |
— |
Step-DeepResearch Technical Report
arXiv preprint arXiv:2512.20491 |
2025-12 |
13 |
GitHub · ★ 575 |
Towards a Science of Scaling Agent Systems
arXiv |
2025-12 |
153 |
GitHub · ★ 63 |
DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems
International Conference on Learning Representations |
2025-12 |
31 |
GitHub · ★ 44 |
PARC: An Autonomous Self-Reflective Coding Agent for Robust Execution of Long-Horizon Tasks
arXiv |
2025-12 |
3 |
— |
Solving a Million-Step LLM Task with Zero Errors
arXiv preprint arXiv:2511.09030 |
2025-11 |
26 |
GitHub · ★ 47 |
LongCodeZip: Compress Long Context for Code Language Models
2025 40th IEEE/ACM International Conference on Automated Software Engineering (ASE) |
2025-10 |
51 |
GitHub · ★ 170 |
ACON: Optimizing Context Compression for Long-horizon LLM Agents
arXiv preprint arXiv:2510.00615 |
2025-10 |
145 |
GitHub · ★ 130 |
Scaling Long-Horizon LLM Agent via Context-Folding
arXiv preprint arXiv:2510.11967 |
2025-10 |
149 |
GitHub · ★ 201 |
AgentFold: Long-Horizon Web Agents with Proactive Context Management
arXiv preprint arXiv:2510.24699 |
2025-10 |
96 |
GitHub · ★ 20,097 |
Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
arXiv |
2025-10 |
453 |
GitHub · ★ 1,423 |
WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research
arXiv preprint arXiv:2509.13312 |
2025-09 |
50 |
GitHub · ★ 20,097 |
ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
arXiv preprint arXiv:2509.13313 |
2025-09 |
141 |
GitHub · ★ 20,097 |
Reducing Cost of LLM Agents with Trajectory Reduction
Proceedings of the ACM on Software Engineering |
2025-09 |
60 |
— |
Where LLM Agents Fail and How They can Learn From Failures
arXiv |
2025-09 |
146 |
GitHub · ★ 108 |
Memp: Exploring Agent Procedural Memory
Findings of the Association for Computational Linguistics: ACL 2026 |
2025-08 |
95 |
GitHub · ★ 41 |
The Complexity Trap: Simple Observation Masking Is as Efficient as LLM Summarization for Agent Context Management
arXiv preprint arXiv:2508.21433 |
2025-08 |
56 |
— |
Magentic-UI: Towards Human-in-the-loop Agentic Systems
arXiv |
2025-07 |
60 |
GitHub · ★ 10,110 |
ReVeal: Self-Evolving Code Agents via Reliable Self-Verification
The Fourteenth International Conference on Learning Representations |
2025-06 |
14 |
— |
SWE-Dev: Building Software Engineering Agents with Training and Inference Scaling
Findings of the Association for Computational Linguistics: ACL 2025 |
2025-06 |
34 |
GitHub · ★ 69 |
MEM1: Learning to Synergize Memory and Reasoning for Efficient Long-Horizon Agents
arXiv preprint arXiv:2506.15841 |
2025-06 |
241 |
GitHub · ★ 353 |
Runaway is Ashamed, But Helpful: On the Early-Exit Behavior of Large Language Model-based Agents in Embodied Environments
Findings of the Association for Computational Linguistics: EMNLP 2025 |
2025-05 |
14 |
GitHub · ★ 2 |
Is there a half-life for the success rates of AI agents?
arXiv preprint arXiv:2505.05115 |
2025-05 |
7 |
— |
Darwin Godel Machine: Open-ended evolution of self-improving agents
arXiv |
2025-05 |
394 |
GitHub · ★ 2,479 |
Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
European Conference on Artificial Intelligence (ECAI) |
2025-04 |
793 |
GitHub · ★ 68,321 |
Process Reward Models That Think
Transactions on Machine Learning Research |
2025-04 |
120 |
GitHub · ★ 93 |
Why Do Multi-Agent LLM Systems Fail?
Advances in Neural Information Processing Systems |
2025-03 |
756 |
GitHub · ★ 433 |
Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents
The Thirteenth International Conference on Learning Representations |
2025-03 |
24 |
GitHub · ★ 7 |
Process Reward Models for LLM Agents: Practical Framework and Directions
arXiv |
2025-02 |
98 |
GitHub · ★ 64 |
A-MEM: Agentic Memory for LLM Agents
Advances in Neural Information Processing Systems |
2025-02 |
1222 |
GitHub · ★ 1,010 |
Practical Considerations for Agentic LLM Systems
arXiv |
2024-12 |
18 |
— |
Godel Agent: A self-referential agent framework for recursive self-improvement
arXiv |
2024-10 |
25 |
GitHub · ★ 239 |
Agent-as-a-Judge: Evaluate Agents with Agents
Forty-second International Conference on Machine Learning |
2024-10 |
269 |
GitHub · ★ 845 |
Agent Workflow Memory
Forty-second International Conference on Machine Learning |
2024-09 |
399 |
GitHub · ★ 481 |
Automated design of agentic systems
ICLR 2025 |
2024-08 |
444 |
GitHub · ★ 1,659 |
LLM Critics Help Catch LLM Bugs
arXiv preprint arXiv:2407.00215 |
2024-07 |
176 |
— |
Demystifying LLM-Based Software Engineering Agents
Proceedings of the ACM on Software Engineering |
2024-07 |
550 |
GitHub · ★ 2,135 |
When Can LLMs Actually Correct Their Own Mistakes? A Critical Survey of Self-Correction of LLMs
Transactions of the Association for Computational Linguistics |
2024-06 |
389 |
— |
HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models
Advances in Neural Information Processing Systems |
2024-05 |
410 |
GitHub · ★ 4,096 |
JARVIS-1: Open-World Multi-Task Agents With Memory-Augmented Multimodal Language Models
IEEE Transactions on Pattern Analysis & Machine Intelligence |
2023-11 |
231 |
GitHub · ★ 416 |
Large Language Models Cannot Self-Correct Reasoning Yet
International Conference on Learning Representations |
2023-10 |
1339 |
— |
MemGPT: Towards LLMs as Operating Systems
arXiv preprint arXiv:2310.08560 |
2023-10 |
1624 |
GitHub · ★ 25,326 |
Self-Taught Optimizer (STOP): Recursively self-improving code generation
Conference on Language Modeling |
2023-10 |
193 |
GitHub · ★ 57 |
Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models
International Conference on Machine Learning |
2023-10 |
714 |
GitHub · ★ 866 |
Cognitive Architectures for Language Agents
Transactions on Machine Learning Research |
2023-09 |
556 |
— |
ExpeL: LLM Agents Are Experiential Learners
AAAI 2024 |
2023-08 |
1144 |
GitHub · ★ 250 |
CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
International Conference on Learning Representations |
2023-05 |
1036 |
GitHub · ★ 746 |
AdaPlanner: Adaptive Planning from Feedback with Language Models
Advances in Neural Information Processing Systems |
2023-05 |
264 |
GitHub · ★ 129 |
Voyager: An Open-Ended Embodied Agent with Large Language Models
Transactions on Machine Learning Research |
2023-05 |
2647 |
GitHub · ★ 7,310 |
Generative Agents: Interactive Simulacra of Human Behavior
Proceedings of the 36th annual acm symposium on user interface software and technology |
2023-04 |
6015 |
GitHub · ★ 22,332 |
Self-Refine: Iterative refinement with self-feedback
NeurIPS 2023 |
2023-03 |
5147 |
GitHub · ★ 847 |
Reflexion: language agents with verbal reinforcement learning
NeurIPS 2023 |
2023-03 |
6117 |
GitHub · ★ 3,394 |
ReAct: Synergizing Reasoning and Acting in Language Models
International Conference on Learning Representations (ICLR) |
2022-10 |
12662 |
GitHub · ★ 4,260 |
-B31B1B.svg?style=flat-square){kind=icon}
