Anagha Gokul, Jason Hartline, Lunjia Hu +2cs.DS cs.GT cs.LG
Calibration requires probabilistic reports to be conditionally unbiased and reliably interpretable as probabilities. A calibration measure assigns numerical error to miscalibrated reports. Haghtalab et al. (2024) proposed an approximately truthful calibration measure for online prediction, leaving open whether exact truthfulness is compatible with completeness and soundness. We resolve this question negatively for sequential binary prediction: exact truthfulness is incompatible with completeness and soundness, even for independent outcomes. We then show that this impossibility is specific to exact truthfulness. We give two general reductions from a base calibration measure, producing additively and multiplicatively approximately truthful calibration measures, respectively. Applying the multiplicative reduction, for every $0 < \varepsilon < 1$ we construct a sound and complete calibration measure that is $(1+\exp(-T^{(1-\varepsilon)/2}/2))$-multiplicatively truthful. This improves the approximate-truthfulness guarantee of Haghtalab et al. (2024).
Representation engineering offers a lightweight means of controlling language-model behavior by modifying intermediate hidden states, but its direct application to Mixture-of-Experts (MoE) models introduces a structural mismatch. We first verify this failure mode through a series of empirical studies and find that preserving clean routing substantially recovers steering performance and that routing is more sensitive to semantic content than to behavioral changes under controlled content. Motivated by these findings, we introduce RARE, a router-agnostic representation engineering framework for MoE language models. RARE projects arbitrary behavioral perturbations onto the null space of the router matrix, thereby removing router-visible components, and further corrects routing drift propagated to selected downstream layers. To decide the best perturbation estimator in this framework, we evaluate five estimators on six heterogeneous open-weight MoE models across three steering scenarios: harmfulness, truthfulness, and factual editing. On harmfulness steering, RARE reaches an average attack success rate of 53.3% while retaining 67.8% MMLU accuracy, yielding a stronger aggregate effectiveness--utility trade-off than baselines. It further improves average TruthfulQA MC1 accuracy from 41.0% to 58.6% and CounterFact efficacy from 16.8% to 96.3%. These results support routing consistency as an important architectural consideration for adapting representation engineering to MoE models.
The Workshop on Trustworthy Natural Language Processing (TrustNLP), co-located with major ACL conferences since 2021, has grown from 8 proceedings papers to 41 over six editions, documenting a field-wide transition from post-hoc interpretability of static models to mechanistic understanding and proactive control of generative systems. We synthesize insights from all 144 proceedings papers, classifying them along six trust dimensions grounded in established frameworks (TrustLLM, DecodingTrust). We observe co-occurrences with capability emergence. The release of the first high-impact chat models activated all trust dimensions simultaneously, while subsequent model generations shifted focus toward truthfulness and safety alignment. Analysis from the classification study reveals that truthfulness is the fastest-growing dimension (absent in 2021-2022, comprising 37% of papers by 2025-2026), fairness remains the most consistent theme, and explainability exhibits a U-shaped trajectory; declining as post-hoc methods lost relevance but resurging in 2026 through mechanistic interpretability. A cross-venue comparison with ACL, NAACL, EACL, and EMNLP (~2K papers) in the same period shows that TrustNLP's topical distribution closely follows the field average. We identify four structural insights and conclude with actionable directions for the research community.
In this work, we introduce CHAIR (Classifier of Hallucination As ImproveR), a supervised framework for detecting hallucinations by analyzing internal logits from each layer of every token. Our method extracts a compact set of features such as maximum, minimum, mean, standard deviation, and slope-from the token logits across all layers, enabling effective hallucination detection without overfitting. Experiments on TruthfulQA and MMLU datasets demonstrate that CHAIR significantly improves detection accuracy, particularly in zero-shot scenarios, showcasing its robustness and generalizability. Beyond hallucination detection, CHAIR highlights the potential of using internal representations for designing advanced decoding strategies. By leveraging patterns in logits, we suggest that more sophisticated models and adaptive decoding methods could further reduce hallucinations and enhance text completion quality. CHAIR not only offers a practical solution for detecting hallucinations but also lays the groundwork for exploring richer representations in LLMs to improve their factuality and coherence.
LLMs have evolved from basic chatbots to the backbone of the AI ecosystem, now widely used in healthcare, schools, and government services. The domain-wide adoption of LLMs necessitates continuous evaluation to ensure their safety and fairness. Common issues encountered after deploying LLMs include inconsistent outputs and hallucinations of incorrect information. Although numerous LLM evaluation tools exist, most are limited to testing a single parameter at a time or require massive computational resources that are not accessible to most researchers. TriEval addresses these challenges by evaluating LLM outputs across multiple parameters, including bias, toxicity, and truthfulness together, while minimizing computing resources. The pipeline is compatible with both open- and closed-source models and runs on a standard laptop without a GPU cluster. TriEval has been tested on four models: Llama 3 8B, Mistral 7B, Gemma 2 9B, and Claude Haiku. The results show clear differences between open-source and closed-source models, especially in terms of toxicity and truthfulness. TriEval is being released as open source to enable broader access for researchers with limited computational resources.
Large Vision-Language Models (VLMs) have achieved remarkable multimodal performance yet remain prone to factual hallucinations, particularly in long-tail or specialized domains. Moreover, current models exhibit a weak capacity to refuse queries that exceed their parametric knowledge. In this paper, we propose a systematic framework to enhance the refusal capability of VLMs when facing such unknown questions. We first curate a model-specific "Visual-Idk" (Visual-I don't know) dataset, leveraging multi-sample consistency probing to distinguish between known and unknown facts. We then align the model using supervised fine-tuning followed by preference-aware optimization (e.g., DPO, ORPO) to effectively delineate its knowledge boundaries. Results on the Visual-Idk dataset show our method improves the Truthful Rate from 57.9\% to 67.3\%. Additionally, internal probing also demonstrates that the model genuinely recognizes its boundaries instead of just memorizing refusal patterns. Our framework further generalizes to out-of-distribution medical and perceptual domains, providing a robust path toward more trustworthy and prudent visual assistants.
As the application of Large Language Models (LLMs) spreads across various industries, there are increasing concerns about the potential for their misuse, especially in sensitive areas such as political discourse. Deliberately aligning LLMs with specific political ideologies, through prompt engineering or fine-tuning techniques, can be advantageous in use cases such as political campaigns, but requires careful consideration due to heightened risks of performance degradation, misinformation, or increased biased behavior. In this work, we propose a multi-dimensional framework inspired by Habermas' Theory of Communicative Action to audit politically aligned language models across four dimensions: effectiveness, fairness, truthfulness, and persuasiveness using automated, quantitative metrics. Applying this to nine popular LLMs aligned via fine-tuning or role-playing revealed consistent trade-offs: while larger models tend to be more effective at role-playing political ideologies and truthful in their responses, they were also less fair, exhibiting higher levels of bias in the form of angry and toxic language towards people of different ideologies. Fine-tuned models exhibited lower bias and more effective alignment than the corresponding role-playing models, but also saw a decline in performance reasoning tasks and an increase in hallucinations. Overall, all of the models tested exhibited some deficiency in at least one of the four metrics, highlighting the need for more balanced and robust alignment strategies. Ultimately, this work aims to ensure politically-aligned LLMs generate legitimate, harmless arguments, offering a framework to evaluate the responsible political alignment of these models.