It Helps to Take a Second Opinion: Teaching Smaller LLMs to Deliberate Mutually via Selective Rationale Optimisation

• URL: http://arxiv.org/abs/2503.02463v1
• Date: Tue, 04 Mar 2025 10:17:29 GMT
• Title: It Helps to Take a Second Opinion: Teaching Smaller LLMs to Deliberate Mutually via Selective Rationale Optimisation
• Authors: Sohan Patnaik, Milan Aggarwal, Sumit Bhatia, Balaji Krishnamurthy,
• Abstract summary: COALITION is a trainable framework that facilitates interaction between two variants of the same SLM.<n>It trains them to generate and refine rationales optimized for the end-task.<n>Our ablation studies reveal that cross-communication between the two variants performs better than using the single model to self-refine the rationales.
• Score: 20.784944581469205
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Very large language models (LLMs) such as GPT-4 have shown the ability to handle complex tasks by generating and self-refining step-by-step rationales. Smaller language models (SLMs), typically with < 13B parameters, have been improved by using the data generated from very-large LMs through knowledge distillation. However, various practical constraints such as API costs, copyright, legal and ethical policies restrict using large (often opaque) models to train smaller models for commercial use. Limited success has been achieved at improving the ability of an SLM to explore the space of possible rationales and evaluate them by itself through self-deliberation. To address this, we propose COALITION, a trainable framework that facilitates interaction between two variants of the same SLM and trains them to generate and refine rationales optimized for the end-task. The variants exhibit different behaviors to produce a set of diverse candidate rationales during the generation and refinement steps. The model is then trained via Selective Rationale Optimization (SRO) to prefer generating rationale candidates that maximize the likelihood of producing the ground-truth answer. During inference, COALITION employs a controller to select the suitable variant for generating and refining the rationales. On five different datasets covering mathematical problems, commonsense reasoning, and natural language inference, COALITION outperforms several baselines by up to 5%. Our ablation studies reveal that cross-communication between the two variants performs better than using the single model to self-refine the rationales. We also demonstrate the applicability of COALITION for LMs of varying scales (4B to 14B parameters) and model families (Mistral, Llama, Qwen, Phi). We release the code for this work at https://github.com/Sohanpatnaik106/coalition.

Related papers

• Scaling Autonomous Agents via Automatic Reward Modeling And Planning [52.39395405893965]
  Large language models (LLMs) have demonstrated remarkable capabilities across a range of tasks.<n>However, they still struggle with problems requiring multi-step decision-making and environmental feedback.<n>We propose a framework that can automatically learn a reward model from the environment without human annotations.
  arXiv  Detail & Related papers  (2025-02-17T18:49:25Z)
• Few-shot Steerable Alignment: Adapting Rewards and LLM Policies with Neural Processes [50.544186914115045]
  Large language models (LLMs) are increasingly embedded in everyday applications.<n> Ensuring their alignment with the diverse preferences of individual users has become a critical challenge.<n>We present a novel framework for few-shot steerable alignment.
  arXiv  Detail & Related papers  (2024-12-18T16:14:59Z)
• Reference Trustable Decoding: A Training-Free Augmentation Paradigm for Large Language Models [79.41139393080736]
  Large language models (LLMs) have rapidly advanced and demonstrated impressive capabilities. In-Context Learning (ICL) and. Efficient Fine-Tuning (PEFT) are currently two mainstream methods for augmenting. LLMs to downstream tasks. We propose Reference Trustable Decoding (RTD), a paradigm that allows models to quickly adapt to new tasks without fine-tuning.
  arXiv  Detail & Related papers  (2024-09-30T10:48:20Z)
• Self-Training with Direct Preference Optimization Improves Chain-of-Thought Reasoning [5.487210426671288]
  In this work, we demonstrate that the reasoning abilities of small-scale LMs can be enhanced through self-training. We also show that the conventional self-training can be further augmented by a preference learning algorithm called Direct Preference Optimization.
  arXiv  Detail & Related papers  (2024-07-25T17:59:16Z)
• Aligning Language Models with Demonstrated Feedback [58.834937450242975]
  Demonstration ITerated Task Optimization (DITTO) directly aligns language model outputs to a user's demonstrated behaviors. We evaluate DITTO's ability to learn fine-grained style and task alignment across domains such as news articles, emails, and blog posts.
  arXiv  Detail & Related papers  (2024-06-02T23:13:56Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)
• Zero-shot Task Preference Addressing Enabled by Imprecise Bayesian Continual Learning [19.11678487931003]
  We propose Imprecise Bayesian Continual Learning (IBCL) to address preferences on task-performance trade-offs. IBCL does not require any additional training overhead to construct preference-addressing models from its knowledge base. We show that models obtained by IBCL have guarantees in identifying the preferred parameters.
  arXiv  Detail & Related papers  (2023-05-24T06:39:00Z)
• A Lagrangian Duality Approach to Active Learning [119.36233726867992]
  We consider the batch active learning problem, where only a subset of the training data is labeled. We formulate the learning problem using constrained optimization, where each constraint bounds the performance of the model on labeled samples. We show, via numerical experiments, that our proposed approach performs similarly to or better than state-of-the-art active learning methods.
  arXiv  Detail & Related papers  (2022-02-08T19:18:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.