Related papers: Distilling Bayesian Belief States into Language Mo…

Language models rely on semantic priors to perform in-context learning, which leads to poor performance on tasks involving inductive reasoning. Instruction-tuning methods based on imitation learning can superficially enhance the in-context…

An ideal dialogue system requires continuous skill acquisition and adaptation to new tasks while retaining prior knowledge. Dialogue State Tracking (DST), vital in these systems, often involves learning new services and confronting…

Large Language Models (LLMs) often exhibit misalignment between the quality of their generated responses and the confidence estimates they assign to them. Bayesian treatments, such as marginalizing over a reliable weight posterior or over…

On-policy distillation (OPD) trains a student on its own trajectories under token-level teacher supervision, but existing methods are capped by a single-teacher capability ceiling: when the teacher errs, the student inherits the error. OPD…

Although pretrained language models (PTLMs) have been shown to contain significant amounts of world knowledge, they can still produce inconsistent answers to questions when probed, even after using specialized training techniques to reduce…

Recent advances in Large Audio-Language Models (LALMs) have made real-time, streaming spoken interaction increasingly practical. In this setting, reasoning quality and responsiveness are tightly coupled: delaying reasoning until the speech…

Distilling knowledge from large proprietary models (e.g., GPT-4) to tiny deployable models (less than 1B parameters) faces a critical capacity-budget trap: the 1000x capacity gap between teachers and students prevents effective direct…

Despite the success of distillation in large language models (LLMs), most prior work applies identical loss functions to both teacher- and student-generated data. These strategies overlook the synergy between loss formulations and data…

Vision-Language Models (VLMs) bring powerful understanding and reasoning capabilities to multimodal tasks. Meanwhile, the great need for capable aritificial intelligence on mobile devices also arises, such as the AI assistant software. Some…

Large language models (LLMs) excel at reasoning tasks but are expensive to deploy. Thus small language models (SLMs) are fine-tuned on CoT data generated by LLMs to copy LLMs' abilities. However, these CoT data may include noisy rationales…

A computationally expensive and memory intensive neural network lies behind the recent success of language representation learning. Knowledge distillation, a major technique for deploying such a vast language model in resource-scarce…

Reasoning-centric large language models (LLMs) achieve strong performance by generating intermediate reasoning trajectories, but often incur excessive token usage and high inference-time decoding cost. We observe that, when solving the same…

Large language models often solve tasks from a fully specified prompt but degrade when the same requirements unfold over multiple turns, known as the lost-in-conversation (LiC) gap. We trace part of this degradation to self-contamination:…

Textual representation learners trained on large amounts of data have achieved notable success on downstream tasks; intriguingly, they have also performed well on challenging tests of syntactic competence. Given this success, it remains an…

Knowledge distillation (KD) is an effective model compression method that can transfer the internal capabilities of large language models (LLMs) to smaller ones. However, the multi-modal probability distribution predicted by teacher LLMs…

Automated negotiation in complex, multi-party and multi-issue settings critically depends on accurate opponent modeling. However, conventional numerical-only approaches fail to capture the qualitative information embedded in natural…

Self-distillation (SD) offers a promising path for adapting large language models (LLMs) without relying on stronger external teachers. However, SD in autoregressive LLMs remains challenging because self-generated trajectories are…

This paper introduces BI-Directional DEliberation Reasoning (BIDDER), a novel reasoning approach to enhance the decision rationality of language models. Traditional reasoning methods typically rely on historical information and employ…

Large language models (LLMs) are increasingly deployed on long-horizon tasks in partially observable environments, where they must act while inferring and tracking a complex environment state over many steps. This leads to two challenges:…

We study self-distillation in settings where supervision is unreliable: there are no ground truth labels, verifiable rewards, or external graders to evaluate answers. We focus on document-grounded question answering with asymmetric context,…