Papers › Can Language Models Explain Their Own Classification Behavior?

Can Language Models Explain Their Own Classification Behavior?

13 May 2024arXiv:2405.07436archive 2025-07-28

Dane Sherburn, Bilal Chughtai, Owain Evans

Large language models (LLMs) perform well at a myriad of tasks, but explaining the processes behind this performance is a challenge. This paper investigates whether LLMs can give faithful high-level explanations of their own internal processes. To explore this, we introduce a dataset, ArticulateRules, of few-shot text-based classification tasks generated by simple rules. Each rule is associated with a simple natural-language explanation. We test whether models that have learned to classify inputs competently (both in- and out-of-distribution) are able to articulate freeform natural language explanations that match their classification behavior. Our dataset can be used for both in-context and finetuning evaluations. We evaluate a range of LLMs, demonstrating that articulation accuracy varies considerably between models, with a particularly sharp increase from GPT-3 to GPT-4. We then investigate whether we can improve GPT-3's articulation accuracy through a range of methods. GPT-3 completely fails to articulate 7/10 rules in our test, even after additional finetuning on correct explanations. We release our dataset, ArticulateRules, which can be used to test self-explanation for LLMs trained either in-context or by finetuning.

PaperPDFCode

Code

danesherbs/articulate-rules officialmentioned in paper report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Classification

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDropoutGPT-3GPT-4Label SmoothingLayer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections