On the Robustness of LLMs' Internal Representation of Code Correctness
Source
Francisco Ribeiro, Sohaila Abdulsattar, Renata Gonzalez, Mahmoud Kassem, Sarah Nadi
Author
Francisco Ribeiro, Sohaila Abdulsattar, Renata Gonzalez, Mahmoud Kassem, Sarah Nadi
Date
Why it matters
Before building verification on internal correctness signals, know the effect looks sensitive to how it is extracted rather than being a stable property of the model — the earlier result does not transfer as a drop-in.
articleThe Knowing-Saying Gap: When Probes See Errors that Confidence MissesJyotin Goel, Ipshita Bandyopadhyay, Justin Shenk
articleValidation Evidence in LLM Repair Agents: How Much of What Passes Actually Tests the Bug?Xiaonan Xu, Wenjing Wu