COFT: Counterfactual-Conformal Decoding for Fair Chain-of-Thought Reasoning in Large Language Models
Large language models (LLMs) can reveal and amplify societal biases during chain-of-thought (CoT) generation. We present COFT (Chain of Fair Thought), a training-free decoding method that applies token-level fairness control at decode time, with distribution-free marginal validity guarantees (under exchangeability) for any frozen causal language model. COFT operates in three stages. First, it creates a masked counterfactual prompt by replacing sensitive spans with neutral tokens. Second, it compares the factual and masked logit distributions through lightweight logit fusion to attenuate attribute-driven biases. Third, it uses dual-branch split-conformal calibration to certify per-step candidate token sets at a user-chosen risk level. We evaluate COFT across six models and multiple bias benchmarks. Our method reduces standard bias metrics by 30-55% (median 38%) while preserving task utility and language quality. Reasoning accuracies remain unchanged within run-to-run noise margins. The computational overhead is modest, equivalent to one additional cached forward pass (<=11%). COFT offers a clear, auditable path to safer CoT generation with significant bias reduction, negligible utility loss, and no requirement for retraining, auxiliary classifiers, or weight access.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Counterfactually Fair Conformal Prediction
While counterfactual fairness of point predictors is well studied, its extension to prediction sets--central to fair decision-making under uncertainty--remains underexplored. On the other hand, conformal prediction (CP) …
Fine-tuning Pre-trained Vision-Language Models in a Human-Annotation-Free Manner
Large-scale vision-language models (VLMs) such as CLIP exhibit strong zero-shot generalization, but adapting them to downstream tasks typically requires costly labeled data. Existing unsupervised self-training methods re…
parameter-efficient fine-tuningZero-shot GeneralizationContrastive LearningSCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation
Accurate representation in media is known to improve the well-being of the people who consume it. Generative image models trained on large web-crawled datasets such as LAION are known to produce images with harmful stere…
Image GenerationSynthetic Counterfactual Labels for Efficient Conformal Counterfactual Inference
This work addresses the problem of constructing reliable prediction intervals for individual counterfactual outcomes. Existing conformal counterfactual inference (CCI) methods provide marginal coverage guarantees but oft…
Coarse-to-Fine Highlighting: Reducing Knowledge Hallucination in Large Language Models
Generation of plausible but incorrect factual information, often termed hallucination, has attracted significant research interest. Retrieval-augmented language model (RALM) -- which enhances models with up-to-date knowl…
HallucinationLanguage ModelingLanguage ModellingQuestion Answering+1