Wrong but Useful: Trajectory Value Beyond Answer Correctness in Multi-Agent Messages
cs.AI, cs.CL, cs.LG
Submitted: 2026-08-14
Updated: 2026-10-31
Terminology
Sources
- Benchmarks Saturate When The Model Gets Smarter Than The Judge
- The Trace Is the State: Exact Credit Assignment for LLM Agent Teams
- Improving Factuality and Reasoning in Language Models through Multiagent Debate
- Gemma 4 Technical Report
- Large Language Models Cannot Self-Correct Reasoning Yet
- Feedback Friction: LLMs Struggle to Fully Incorporate External Feedback
- LAB-Bench: Measuring Capabilities of Language Models for Biology Research
- Let's Verify Step by Step
- CriticBench: Benchmarking LLMs for Critique-Correct Reasoning
- A Dynamic LLM-Powered Agent Network for Task-Oriented Agent Collaboration
- Agents that Matter: How Subtle Choices Shape Agent Attribution in Multi-Agent Systems
- Self-Refine: Iterative Refinement with Self-Feedback
- gpt-oss-120b & gpt-oss-20b Model Card
- Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failures
- Reflexion: Language Agents with Verbal Reinforcement Learning
- MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning
- Multi-Agent Collaboration Mechanisms: A Survey of LLMs
- Solving math word problems with process- and outcome-based feedback
- Can Large Language Models Really Improve by Self-critiquing Their Own Plans?
- Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents