Agent-Reflex: Teaching Language Agents to Act Reflectively
cs.CL
Submitted: 2026-10-13
Updated: 2026-10-13
Comments: Working in progress
Code: https://github.com/RUC-NLPIR/Agent-Reflex
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoning
- ACEBench: Who Wins the Match Point in Tool Usage?
- Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments
- Agentic Entropy-Balanced Policy Optimization
- Tool-Star: Empowering LLM-Brained Multi-Tool Reasoner via Reinforcement Learning
- Agentic Reinforced Policy Optimization
- Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
- The Llama 3 Herd of Models
- GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators
- VitaBench: Benchmarking LLM Agents with Versatile Interactive Tasks in Real-world Applications
- REINFORCE++: Stabilizing Critic-Free Policy Optimization with Global Advantage Normalization
- SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks
- Simulating Environments with Reasoning Models for Agent Training
- JobBench: Aligning Agent Work With Human Will
- Self-Refine: Iterative Refinement with Self-Feedback
- gpt-oss-120b & gpt-oss-20b Model Card
- OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning
- GDPval: Evaluating AI Model Performance on Real-World Economically Valuable Tasks
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering