When Tools Hurt LLM Reasoning: State-Dependent Belief Revision under External Evidence
cs.CL, cs.AI
Submitted: 2025-08-21
Updated: 2026-09-07
Comments: Accepted by EMNLP 2026
Code: https://github.com/epsilondylan/State-Dependent-Belief-Revision
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Towards Effective Code-Integrated Reasoning
- Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Cost-of-Pass: An Economic Framework for Evaluating Language Models
- ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
- Training Large Language Models to Reason in a Continuous Latent Space
- A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems
- ToRL: Scaling Tool-Integrated RL
- THINK-Bench: Evaluating Thinking Efficiency and Chain-of-Thought Quality of Large Reasoning Models
- ZebraLogic: On the Scaling Limits of LLMs for Logical Reasoning
- Deciphering Trajectory-Aided LLM Reasoning: An Optimization Perspective
- Are Your LLMs Capable of Stable Reasoning?
- s1: Simple test-time scaling
- Concise Thoughts: Impact of Output Length on LLM Reasoning and Cost
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Agentic Reasoning and Tool Integration for LLMs via Reinforcement Learning
- Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
- Kimi k1.5: Scaling Reinforcement Learning with LLMs
- Rethinking Inference-Time Scaling: Efficiency Limits and Linguistic Signals
- Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs
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