The thesis explores Introspective Uncertainty Estimation (IUE) for large language models (LLMs) in code generation, aiming to determine whether hidden-state representations can indicate functional correctness at both response and line levels. Using LiveCodeBench and BigCodeBench, the study finds that hidden states provide a strong signal for overall correctness, with static single-token probes performing best, while dynamic strategies offer no consistent advantage. Although line-level fault localization is more challenging, a conditional Top‑K ranking approach remains effective, suggesting a two‑stage workflow that first screens responses for risk and then prioritizes line‑level checks.
By Thomas Klassert
arXiv:2606. 17514v1 Announce Type: cross Abstract: Large Language Models have shown remarkable capabilities in code generation.
By Le Zhang, Suresh Kothari
arXiv:2609.39568v1 Announce Type: cross
Abstract: Large language models (LLMs) may generate unreliable code on corner cases missed by testing, while formal verification can provide machine-checkable...
By Jiaru Qian, Yihong Dong, Yongmin Li, Hao Zhu, Bin Gu, Ge Li
arXiv:2608. 14659v1 Announce Type: new Abstract: Large language models for code generation often produce incorrect solutions without reliable indicators of failure.
By Pranav Rakasi, Maanas Lalwani, Arnav Srivastava, Arya Palanivel, Tinuade Adeleke, Ruizhe Li, Sean Wu
arXiv:2610.00425v1 Announce Type: cross
Abstract: Code generation has emerged as a central capability of large language models, with coding agents now able to produce functionally correct software pr...
By Bhanu Prakash Vangala, Tanu Malik
The paper investigates how to improve confidence calibration for large language models (LLMs) used in automated code revision (ACR). It proposes applying local Platt-scaling to three fine-grained confidence scores, rather than the conventional global method, and demonstrates that this approach consistently reduces calibration error across multiple tasks, metrics, and model sizes. The study shows that fine-grained calibration, especially when combined with global scaling, yields more reliable confidence estimates for ACR tasks.
By Hong Yi Lin, Chunhua Liu, Haoyu Gao, Patanamon Thongtanunam, Christoph Treude
arXiv:2606. 00920v1 Announce Type: cross Abstract: Run-level pass rate overstates retry-free coverage by up to 17.
By Yongxi Zhou, Lai Yun Choi, Jiaxi Wen, Wenbo Ye
arXiv:2608. 08266v1 Announce Type: cross Abstract: Code generated by modern language models often reads naturally.
By Francisco Ribeiro, Sohaila Abdulsattar, Renata Gonzalez, Mahmoud Kassem, Sarah Nadi
arXiv:2505. 13553v3 Announce Type: replace-cross Abstract: The hallucination of code generation models hinders their applicability to systems requiring higher safety standards.
By Jaewoo Jeong, Taesoo Kim, Sangdon Park
arXiv:2511. 20709v2 Announce Type: replace-cross Abstract: Large language models (LLMs) and LLM-based coding agents are now used to generate code from natural-language specifications, yet ensuring such code is both functionally correct and secure remains a challenge.
By Rupam Patir, Keyan Guo, Suvadra Barua, Abhijeet Pathak, Dinesh Gudimetla, Jiawei Guo, Hongxin Hu, Haipeng Cai
Code language models are now trusted collaborators in production workflows for debugging, refactoring, and iterative repair, and every benchmark that evaluates them assumes the instructions they act on are correct. We study what happens when that assumption breaks.
arXiv:2607. 04537v1 Announce Type: cross Abstract: Code language models are now trusted collaborators in production workflows for debugging, refactoring, and iterative repair, and every benchmark that evaluates them assumes the instructions they act on are correct.
By Raj Jaiswal, Anany Singh Divy, Savar Bhasin, Adi Bajpai, Tanuja Ganu, Rajiv Ratn Shah