Advancing LLM Reasoning Generalists with Preference Trees

• URL: http://arxiv.org/abs/2404.02078v1
• Date: Tue, 2 Apr 2024 16:25:30 GMT
• Title: Advancing LLM Reasoning Generalists with Preference Trees
• Authors: Lifan Yuan, Ganqu Cui, Hanbin Wang, Ning Ding, Xingyao Wang, Jia Deng, Boji Shan, Huimin Chen, Ruobing Xie, Yankai Lin, Zhenghao Liu, Bowen Zhou, Hao Peng, Zhiyuan Liu, Maosong Sun,
• Abstract summary: We introduce Eurus, a suite of large language models (LLMs) optimized for reasoning. Eurus models achieve state-of-the-art results among open-source models on a diverse set of benchmarks.
• Score: 119.57169648859707
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: We introduce Eurus, a suite of large language models (LLMs) optimized for reasoning. Finetuned from Mistral-7B and CodeLlama-70B, Eurus models achieve state-of-the-art results among open-source models on a diverse set of benchmarks covering mathematics, code generation, and logical reasoning problems. Notably, Eurus-70B beats GPT-3.5 Turbo in reasoning through a comprehensive benchmarking across 12 tests covering five tasks, and achieves a 33.3% pass@1 accuracy on LeetCode and 32.6% on TheoremQA, two challenging benchmarks, substantially outperforming existing open-source models by margins more than 13.3%. The strong performance of Eurus can be primarily attributed to UltraInteract, our newly-curated large-scale, high-quality alignment dataset specifically designed for complex reasoning tasks. UltraInteract can be used in both supervised fine-tuning and preference learning. For each instruction, it includes a preference tree consisting of (1) reasoning chains with diverse planning strategies in a unified format, (2) multi-turn interaction trajectories with the environment and the critique, and (3) pairwise data to facilitate preference learning. UltraInteract allows us to conduct an in-depth exploration of preference learning for reasoning tasks. Our investigation reveals that some well-established preference learning algorithms may be less suitable for reasoning tasks compared to their effectiveness in general conversations. Inspired by this, we derive a novel reward modeling objective which, together with UltraInteract, leads to a strong reward model.

Related papers

• SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models [88.29990536278167]
  We introduce SPaR, a self-play framework integrating tree-search self-refinement to yield valid and comparable preference pairs free from distractions. Our experiments show that a LLaMA3-8B model, trained over three iterations guided by SPaR, surpasses GPT-4-Turbo on the IFEval benchmark without losing general capabilities.
  arXiv  Detail & Related papers  (2024-12-16T09:47:43Z)
• MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale [66.73529246309033]
  multimodal large language models (MLLMs) have shown significant potential in a broad range of multimodal tasks. Existing instruction-tuning datasets only provide phrase-level answers without any intermediate rationales. We introduce a scalable and cost-effective method to construct a large-scale multimodal instruction-tuning dataset with rich intermediate rationales.
  arXiv  Detail & Related papers  (2024-12-06T18:14:24Z)
• Preference Optimization for Reasoning with Pseudo Feedback [100.62603571434167]
  We introduce a novel approach to generate pseudo feedback for reasoning tasks by framing the labeling of solutions as an evaluation against associated test cases. We conduct experiments on both mathematical reasoning and coding tasks using pseudo feedback for preference optimization, and observe improvements across both tasks.
  arXiv  Detail & Related papers  (2024-11-25T12:44:02Z)
• Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment [57.03947082589616]
  Large Language Models (LLMs) are often aligned using contrastive alignment objectives and preference pair datasets. We study this and find that preference data gives a better learning signal when the underlying responses are contrastive. We introduce Contrastive Learning from AI Revisions (CLAIR), a data-creation method which leads to more contrastive preference pairs. Our best model, trained on 32K CLAIR preferences with APO, improves Llama-3-8B-Instruct by 7.65%, closing the gap with GPT4-turbo by 45%.
  arXiv  Detail & Related papers  (2024-08-12T16:24:51Z)
• Insights into Alignment: Evaluating DPO and its Variants Across Multiple Tasks [35.36615140853107]
  This study evaluates Direct Preference Optimization (DPO) and its variants for aligning Large Language Models (LLMs) with human preferences. Our evaluation spans 13 benchmarks covering dialogue, reasoning, mathematical problem-solving, question answering, truthfulness, MT-Bench, Big Bench, and the Open LLM Leaderboard. We find that alignment methods often achieve near optimal performance even with smaller subsets of training data.
  arXiv  Detail & Related papers  (2024-04-23T03:55:01Z)
• A Three-Phases SFT Hybrid Model Integrated Strong Prior Module and Data Overlap Estimation in the Eduation Context [0.0]
  We propose an end-to-end prior-based three-phases supervised fine-tuned model. Our model realizes the structural disassembly and incremental guided output of educational knowledge. Our model also achieves state-of-the-art in code abilities compared to open-source models.
  arXiv  Detail & Related papers  (2024-03-13T05:38:39Z)
• AQA-Bench: An Interactive Benchmark for Evaluating LLMs' Sequential Reasoning Ability [29.1826948551409]
  AQA-Bench is a novel benchmark to assess the sequential reasoning capabilities of large language models. We build AQA-Bench with three different algorithms, namely binary search, depth-first search, and breadth-first search. Our investigations reveal several interesting findings.
  arXiv  Detail & Related papers  (2024-02-14T18:59:33Z)
• Tool-Augmented Reward Modeling [58.381678612409]
  We propose a tool-augmented preference modeling approach, named Themis, to address limitations by empowering RMs with access to external environments. Our study delves into the integration of external tools into RMs, enabling them to interact with diverse external sources. In human evaluations, RLHF trained with Themis attains an average win rate of 32% when compared to baselines.
  arXiv  Detail & Related papers  (2023-10-02T09:47:40Z)

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.