mlsanityData to 14 Oct 2026

Capability Self-Assessment in Large Language Models

arXiv 2606.00251 ↗First posted 29 May 2026cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

The ability to recognize one's own limitations and decide whether to solve a problem or seek help is fundamental to reliable intelligent systems. Yet modern large language models tend to overestimate their competence and attempt queries they cannot solve. We study how models can learn Capability Self-Assessment (CSA) while preserving their problem-solving ability. Because the same model both assesses and solves queries, self-assessment training can change the capability it must recognize. We systematically compare supervised fine-tuning (SFT) and reinforcement learning (RL), evaluating each trained model's self-assessment against its updated capability and measuring how well it retains its solving ability. Across model families, scales, and domains, RL provides the strongest assessment-retention trade-off. We also find that CSA transfers to the out-of-distribution settings we evaluate, and demonstrate its practical value for local-cloud routing at inference time and targeted data selection during training.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 73 of its 94 references that are in the map. First posted on arXiv 29 May 2026.

Builds on

73 of 94 references.

References · 73

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  5. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  6. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  7. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  8. 2017On Calibration of Modern Neural Networkscited by 4,960
  9. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  10. 2023Let's Verify Step by Stepcited by 3,789
  11. 2021Program Synthesis with Large Language Modelscited by 3,637
  12. 2022Survey of Hallucination in Natural Language Generationcited by 2,075
38 more
  1. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
  2. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  3. 2021WebGPT: Browser-assisted question-answering with human feedbackcited by 1,745
  4. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  5. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,391
  6. 2023Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMscited by 974
  7. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 853
  8. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 812
  9. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 810
  10. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 682
  11. 2025SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-trainingcited by 672
  12. 2023The Internal State of an LLM Knows When It's Lyingcited by 659
  13. 2022MedMCQA : A Large-scale Multi-Subject Multi-Choice Dataset for Medical domain Question Answeringcited by 543
  14. 2023DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretrainingcited by 399
  15. 2025ReTool: Reinforcement Learning for Strategic Tool Use in LLMscited by 392
  16. 2024Hybrid LLM: Cost-Efficient and Quality-Aware Query Routingcited by 372
  17. 2019SelectiveNet: A Deep Neural Network with an Integrated Reject Optioncited by 350
  18. 20242 OLMo 2 Furiouscited by 340
  19. 2024OLMoE: Open Mixture-of-Experts Language Modelscited by 301
  20. 2024Omni-MATH: A Universal Olympiad Level Mathematic Benchmark For Large Language Modelscited by 284
  21. 2024Semantic Entropy Probes: Robust and Cheap Hallucination Detection in LLMscited by 259
  22. 2023Do Large Language Models Know What They Don't Know?cited by 243
  23. 2022Prioritized Training on Points that are Learnable, Worth Learning, and Not Yet Learntcited by 236
  24. 2024A Survey on Data Selection for Language Modelscited by 236
  25. 2017Predict Responsibly: Improving Fairness and Accuracy by Learning to Defercited by 233
  26. 2024Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?cited by 232
  27. 2023MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Usecited by 217
  28. 2020Reducing conversational agents' overconfidence through linguistic calibrationcited by 210
  29. 2025RL's Razor: Why Online Reinforcement Learning Forgets Lesscited by 176
  30. 2023Conformal Language Modelingcited by 170
  31. 2023Conformal Prediction With Conditional Guaranteescited by 152
  32. 2023Conformal Prediction with Large Language Models for Multi-Choice Question Answeringcited by 139
  33. 2023AutoMix: Automatically Mixing Language Modelscited by 125
  34. 2024AnyTool: Self-Reflective, Hierarchical Agents for Large-Scale API Callscited by 124
  35. 2023On the Tool Manipulation Capability of Open-source Large Language Modelscited by 122
  36. 2024Know Your Limits: A Survey of Abstention in Large Language Modelscited by 121
  37. 2025Beyond Binary Rewards: Training LMs to Reason About Their Uncertaintycited by 117
  38. 2024Looking Inward: Language Models Can Learn About Themselves by Introspectioncited by 113

50 most cited of 73 All 73 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Are We There Yet? Assessing Computer-Use Agents for Blind Users' Accessible Interaction with Desktop Applicationscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025The Path of Self-Evolving Large Language Models: Achieving Data-Efficient Learning via Intrinsic Feedback
  2. 2026ReAD: Reinforcement-Guided Capability Distillation for Large Language Models
  3. 2025KnowRL: Teaching Language Models to Know What They Know
  4. 2025Learning What Reinforcement Learning Can't: Interleaved Online Fine-Tuning for Hardest Questions
  5. 2025Post-Training Large Language Models via Reinforcement Learning from Self-Feedback
  6. 2026Supervised Fine-Tuning versus Reinforcement Learning: A Study of Post-Training Methods for Large Language Models
  7. 2026DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training
  8. 2024Can Large Language Models Invent Algorithms to Improve Themselves?: Algorithm Discovery for Recursive Self-Improvement through Reinforcement Learning