Related papers: Reasoning Distillation for Lightweight Automated P…

Deep learning techniques have achieved great success in many fields, while at the same time deep learning models are getting more complex and expensive to compute. It severely hinders the wide applications of these models. In order to…

Advances in large language models (LLMs) significantly enhance reasoning capabilities but their deployment is restricted in resource-constrained scenarios. Knowledge distillation addresses this by transferring knowledge from powerful…

The shift toward interacting with frozen, "black-box" Large Language Models (LLMs) has transformed prompt engineering from a heuristic exercise into a critical optimization challenge. We propose a Reinforcement Learning (RL) framework for…

With the rapid development of deep learning, large language models have shown strong capabilities in complex reasoning tasks such as mathematical equation solving. However, their substantial computational and storage costs hinder practical…

Model distillation typically focuses on behavioral mimicry, where a student model is trained to replicate a teacher's output while treating its internal computations as a black box. In this work we propose an alternative approach:…

Deep neural networks with millions of parameters may suffer from poor generalization due to overfitting. To mitigate the issue, we propose a new regularization method that penalizes the predictive distribution between similar samples. In…

Evaluating the trustworthiness of a model's prediction is essential for differentiating between `right for the right reasons' and `right for the wrong reasons'. Identifying textual spans that determine the target label, known as faithful…

As the size of pre-trained speech recognition models increases, running these large models in low-latency or resource-constrained environments becomes challenging. In this work, we leverage pseudo-labelling to assemble a large-scale…

Language models (LMs) have yielded impressive results on many language reasoning tasks, but their unexpected errors raise doubts about their reasoning abilities. In light of this, there is growing interest in finetuning/prompting LMs with…

Large language models make remarkable progress in reasoning capabilities. Existing works focus mainly on deductive reasoning tasks (e.g., code and math), while another type of reasoning mode that better aligns with human learning, inductive…

Small Language Models (SLMs) are becoming increasingly popular in specialized fields, such as industrial applications, due to their efficiency, lower computational requirements, and ability to be fine-tuned for domain-specific tasks,…

The outpouring of various pre-trained models empowers knowledge distillation by providing abundant teacher resources, but there lacks a developed mechanism to utilize these teachers adequately. With a massive model repository composed of…

Distillation efforts have led to language models that are more compact and efficient without serious drops in performance. The standard approach to distillation trains a student model against two objectives: a task-specific objective (e.g.,…

Complex deep learning models now achieve state of the art performance for many document retrieval tasks. The best models process the query or claim jointly with the document. However for fast scalable search it is desirable to have document…

Recent advancements have demonstrated that the performance of large language models (LLMs) can be significantly enhanced by scaling computational resources at test time. A common strategy involves generating multiple Chain-of-Thought (CoT)…

Large Language Models (LLMs) excel in reasoning tasks through Chain-of-Thought (CoT) prompting. However, CoT prompting greatly increases computational demands, which has prompted growing interest in distilling CoT capabilities into Small…

Current Natural Language Inference (NLI) models achieve impressive results, sometimes outperforming humans when evaluating on in-distribution test sets. However, as these models are known to learn from annotation artefacts and dataset…

Knowledge distillation (KD) is widely used for training a compact model with the supervision of another large model, which could effectively improve the performance. Previous methods mainly focus on two aspects: 1) training the student to…

Knowledge distillation is typically realized by transferring a teacher model's knowledge into a student's parameters through supervised or reinforcement-based optimization. While effective, such approaches require repeated parameter updates…

On-policy self-distillation (OPSD) improves the reasoning performance of large language models (LLMs) by providing dense token-level supervision for on-policy rollouts. However, existing OPSD methods often yield limited gains on in-domain…