cd /news/large-language-models/what-llms-explain-is-not-what-they-b… · home topics large-language-models article
[ARTICLE · art-44348] src=arxiv.org ↗ pub= topic=large-language-models verified=true sentiment=· neutral

What LLMs explain is not what they believe: Evaluating explanation sufficiency under models' own input beliefs

Researchers introduced SCSuff, an information-theoretic metric to evaluate whether free-text explanations from large language models (LLMs) are sufficient to explain their outputs. Testing on LLMs showed explanations are generally insufficient and weakly correlated with model size or accuracy, with sufficiency varying based on input distribution. The metric can guide detection and improvement of sufficient explanations.

read1 min views1 publishedJun 30, 2026

arXiv:2606.28615v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in high-stakes domains, where free-text explanations such as chain-of-thought and post-hoc rationales are used to justify model outputs. Yet it remains unclear whether these explanations are sufficient, i.e., if they contain enough information to explain the model's output-generating process. We generalize classical sufficiency from feature attributions to arbitrary explanations and prove that explanation sufficiency can change depending on the input distribution, which must be explicitly defined for LLM explanations. We propose using the LLM itself to generate alternative inputs conditioned on an explanation, capturing its beliefs about possible inputs. We formalize self-consistent sufficiency as a goal for free-text explanations and introduce an information-theoretic metric, SCSuff, that enables evaluation of free-text explanations without relying on predefined biases or shortcuts. Our experiments show that SCSuff agrees with targeted perturbation tests where applicable and demonstrate that explanation sufficiency can vary with the input distribution. We find LLM explanations are generally insufficient and weakly correlated with model size, accuracy, or output entropy. Analysis of final-token hidden states shows that top and bottom SCSuff scores can be predicted from internal representations, suggesting that SCSuff can guide detection and improvement of sufficient LLM explanations. The code for this paper is available at https://github.com/rajesh-lab/self-consistent-sufficiency .

── more in #large-language-models 4 stories · sorted by recency
── more on @arxiv 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/what-llms-explain-is…] indexed:0 read:1min 2026-06-30 ·