The paper introduces Context-Dependent Argumentation Frameworks (CDAFs) to model how an agent strategically manipulates the success of arguments by choosing the external evaluation context.

Introduces Posterior Prefix Tuning (PPT) for eliciting high-utility continuations from Bayes-filtered transformers using a latent posterior model.

The paper introduces the Triangulated Preference Shift score, an automated, curation-free metric to quantify systematic lexical biases introduced into Large Language Models during the preference-learn…

The paper provides a formal statistical and conceptual framework for defining and measuring 'pairwise reference alignment,' which quantifies how well a model's scoring function agrees with a given ref…

This paper investigates the effectiveness of stage-dependent preference elicitation strategies in conversational recommendation systems and introduces COPE, a novel architecture for strategy modeling.

The paper introduces PRISM, a method for decoding preference signals from noisy latents using a lightweight Query-based Aggregation head and a frozen video diffusion backbone, achieving state-of-the-a…

The paper proposes Reasoning-Conditioned Direct Preference Optimization (RC-DPO) to effectively mitigate hallucinations in multimodal large reasoning models by explicitly conditioning the preference o…

The paper introduces PARL, a framework that learns personalized evaluation rubrics directly from raw user interaction histories to accurately assess how well LLM outputs align with subjective, user-sp…

This paper analyzes Best-of-$N$ preference data, deriving explicit reward targets for independent-reference variants and establishing design principles for choosing $N$ and the base distribution to op…

The paper introduces CERA, a novel contrastive retrieval framework that improves RAG factuality and interpretability by using subjectivity-based hard negative selection and an auxiliary attention alig…

The paper proposes In-Context Reward Adaptation, a transformer-based framework that uses in-context learning and auxiliary signals (like human response time) to robustly model diverse and unseen human…

This paper proposes HPRO, a hierarchical progressive reward optimization framework for Large Language Model-based Text-to-Speech, to overcome the information conflict and scale gap issues.

The study demonstrates that conditioning AI brand recommendations on a user's persona significantly alters the recommended product set, particularly for mid-market brands, and this effect is largest o…

The paper demonstrates that explicit gender cues systematically affect LLM value trade-offs, causing decision flips that are often masked or misattributed by the models themselves.

The paper proposes WEINCE, a modified InfoNCE objective that uses extreme value theory corrections to improve contrastive learning by more accurately modeling the selection of hard negative examples.

The paper demonstrates that the valence structure learned by modern LLMs aligns with human EEG emotional representations, but finds that further supervised alignment is ineffective due to a phenomenon…

The study found that while contextualizing AI responses reduces their persuasive power, combining this technique with conversational warmth restores persuasiveness, suggesting that user deference to A…