Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning
Xinyan Guan, Jiali Zeng, Chunlei Xin, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun, Fandong Meng
cs.CL
Submitted: 2026-07-31
Code: https://github.com/icip-cas/Knowing-When-to-Quit
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Is Bigger and Deeper Always Better? Probing LLaMA Across Scales and Layers
- Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
- Can AI Assistants Know What They Don't Know?
- SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Measuring Progress on Scalable Oversight for Large Language Models
- Prioritization First, Principles Second: An Adaptive Interpretation of Helpful, Honest, and Harmless Principles
- SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models
- Beyond Two-Stage Training: Cooperative SFT and RL for LLM Reasoning
- Training LLMs for Honesty via Confessions
- Language Models (Mostly) Know What They Know
- Teaching Models to Express Their Uncertainty in Words
- DeepSeek-V3 Technical Report
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
- Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment
- Large Language Models: A Survey
- gpt-oss-120b & gpt-oss-20b Model Card
- A Comprehensive Survey of Hallucination in Large Language, Image, Video and Audio Foundation Models
- Towards Large Reasoning Models: A Survey of Reinforced Reasoning with Large Language Models
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