Searcharxiv⌕ Search

arXiv subjects

Yiliu Sun

Publications and source records attributed to Yiliu Sun.

7 recordsLinked to original sources

Harmony in Diversity: Multi-domain Contrastive Policy Optimization for Large Reasoning Models

Multi-domain capabilities are critical for practically applicable Large Reasoning Models (LRMs). Therefore, mixed-domain Reinforcement learning (RL) has become a crucial process to jointly train a single LRM with mixed domain data, but it often induces capability trade-offs among different domains. Existing methods attribute this issue to competing cross-domain interactions, and mitigate them by isolating learning directions and balancing learning progress. However, our observation suggests that they also impede supportive knowledge sharing across domains, which make it difficult to match or surpass single-domain performance. To address this problem, we propose \textbf{M}ulti-domain \textbf{C}ontrastive \textbf{P}olicy \textbf{O}ptimization (MCPO), which applies contrastive learning to achieve diverse knowledge sharing and competition. We first localize the interaction units on critical reasoning segments, and then formulate mutually augmented examples to capture both positive and negative interactions. Subsequently, we design complementary contrastive objectives for cross-domain knowledge sharing and intra-domain knowledge consolidation. Overall, we utilize diverse cross-domain interactions to achieve compatibility across domains and discriminability within each domain to form a harmonious reasoning space. Experimental results across a broad range of domains show that MCPO significantly outperforms existing methods mitigating interactions over 3.17 average points, and even surpasses single-domain training over 1.05 average points.

cs.CL↗

Janus-MOPD: Tracing the Student Learning Dynamics in Multi-Teacher On-Policy Distillation

Multi-domain capabilities are critical for practically applicable Large Reasoning Models (LRMs). Therefore, Multi-teacher On-Policy Distillation (MOPD) has become a crucial part of the post-training process, integrating the capabilities of multiple domain-specific teacher models into a single student model. However, the student usually fails to acquire multi-domain capabilities simultaneously due to undesirable domain heterogeneity. Existing works have address this problem in view of temporal uneven efficiency and spatial hard allocation in external multi-teacher supervision. While in broader multi-domain training, varying degrees of performance degradation is still largely unexplained. In this paper, we shift the focus on how the student learns and integrates multi-domain capabilities internally, and then propose \textbf{\textsc{Janus-MOPD}}, a general learning dynamics framework of geometric representations for LRMs. Specifically, our main focus is on the internal performance of multi-domain capabilities on both temporal and spatial dimensions within the student. Temporally, our study shows that the student exhibits a Expansion-Consolidation-Optimization (ECO) three-phase learning dynamics, together with inherent limitations in learning efficiency and scaling. Spatially, we summarize two types of subspace locking within and across domains by tracing the representation space of the student, while explaining its clear learning directions but inherent cross-domain conflict. Therefore, \textsc{Janus-MOPD} yields a unified learning dynamics formulation, a geometric tracing methodology, and systematic findings of how the student learns and integrates multi-domain capabilities. Overall, we provide a \textsc{Janus} perspective to internally trace the continual evolution of the multi-domain capabilities within the model itself.

cs.CL↗

TL-GRPO: Turn-Level RL for Reasoning-Guided Iterative Optimization

Large language models have demonstrated strong reasoning capabilities in complex tasks through tool integration, which is typically framed as a Markov Decision Process and optimized with trajectory-level RL algorithms such as GRPO. However, a common class of reasoning tasks, iterative optimization, presents distinct challenges: the agent interacts with the same underlying environment state across turns, and the value of a trajectory is determined by the best turn-level reward rather than cumulative returns. Existing GRPO-based methods cannot perform fine-grained, turn-level optimization in such settings, while black-box optimization methods discard prior knowledge and reasoning capabilities. To address this gap, we propose Turn-Level GRPO (TL-GRPO), a lightweight RL algorithm that performs turn-level group sampling for fine-grained optimization. We evaluate TL-GRPO on analog circuit sizing (ACS), a challenging scientific optimization task requiring multiple simulations and domain expertise. Results show that TL-GRPO outperforms standard GRPO and Bayesian optimization methods across various specifications. Furthermore, our 30B model trained with TL-GRPO achieves state-of-the-art performance on ACS tasks under same simulation budget, demonstrating both strong generalization and practical utility.

cs.CL↗

Well Begun, Half Done: Reinforcement Learning with Prefix Optimization for LLM Reasoning

Reinforcement Learning with Verifiable Rewards (RLVR) significantly enhances the reasoning capability of Large Language Models (LLMs). Current RLVR approaches typically conduct training across all generated tokens, but neglect to explore which tokens (e.g., prefix tokens) actually contribute to reasoning. This uniform training strategy spends substantial effort on optimizing low-return tokens, which in turn impedes the potential improvement from high-return tokens and reduces overall training effectiveness. To address this issue, we propose a novel RLVR approach called Progressive Prefix-token Policy Optimization (PPPO), which highlights the significance of the prefix segment of generated outputs. Specifically, inspired by the well-established human thinking theory of Path Dependence, where early-stage thoughts substantially constrain subsequent thinking trajectory, we identify an analogous phenomenon in LLM reasoning termed Beginning Lock-in Effect (BLE). PPPO leverages this finding by focusing its optimization objective on the prefix reasoning process of LLMs. This targeted optimization strategy can positively influence subsequent reasoning processes, and ultimately improve final results. To improve the learning effectiveness of LLMs on how to start reasoning with high quality, PPPO introduces two training strategies: (a) Progressive Prefix Retention, which shapes a progressive learning process by increasing the proportion of retained prefix tokens during training; (b) Continuation Accumulated Reward, which mitigates reward bias by sampling multiple continuations for one prefix token sequence, and accumulating their scores as the reward signal. Extensive experimental results on various reasoning tasks demonstrate that our proposed PPPO outperforms representative RLVR methods, with the accuracy improvements of 18.02% on only 26.17% training tokens.

cs.CL↗

CortexDebate: Debating Sparsely and Equally for Multi-Agent Debate

Nowadays, single Large Language Model (LLM) struggles with critical issues such as hallucination and inadequate reasoning abilities. To mitigate these issues, Multi-Agent Debate (MAD) has emerged as an effective strategy, where LLM agents engage in in-depth debates with others on tasks. However, existing MAD methods face two major issues: (a) too lengthy input contexts, which causes LLM agents to get lost in plenty of input information and experiences performance drop; and (b) the overconfidence dilemma, where self-assured LLM agents dominate the debate, leading to low debating effectiveness. To address these limitations, we propose a novel MAD method called "CortexDebate". Inspired by the human brain's tendency to establish a sparse and dynamically optimized network among cortical areas governed by white matter, CortexDebate constructs a sparse debating graph among LLM agents, where each LLM agent only debates with the ones that are helpful to it. To optimize the graph, we propose a module named McKinsey-based Debate Matter (MDM), which acts as an artificial analog to white matter. By integrating the McKinsey Trust Formula, a well-established measure of trustworthiness from sociology, MDM enables credible evaluations that guide graph optimization. The effectiveness of our CortexDebate has been well demonstrated by extensive experimental results across eight datasets from four task types.

cs.AI↗

Fast-Slow-Thinking: Complex Task Solving with Large Language Models

Nowadays, Large Language Models (LLMs) have been gradually employed to solve complex tasks. To face the challenge, task decomposition has become an effective way, which proposes to divide a complex task into multiple simpler subtasks and then solve them separately so that the difficulty of the original task can be reduced. However, the performance of existing task decomposition methods can be suboptimal when the task contains overly complex logic and constraints. In this situation, the solution generated by LLMs may deviate from the original purpose of the task, or contain redundant or even erroneous content. Therefore, inspired by the fact that humans possess two thinking systems including fast thinking and slow thinking, this paper introduces a new task decomposition method termed ``Fast-Slow-Thinking'' (FST), which stimulates LLMs to solve tasks through the cooperation of Fast Thinking (FT) and Slow Thinking (ST) steps. Here FT focuses more on the general and concise aspect of the task, and ST focuses more on the details of the task. In FT, LLMs are prompted to remove the constraints of the original task, therefore simplifying it to a general and concise one. In ST, we recall the constraints removed in FT, so that LLMs can improve the answer generated in FT to meet the requirements of the original task. Therefore, our FST method enables LLMs to consider a complex problem via a human-like cognition process from coarse to fine, the effectiveness of which has been well demonstrated by the experiments on three types of tasks.

cs.CL↗

Large Language Models as an Indirect Reasoner: Contrapositive and Contradiction for Automated Reasoning

Recently, increasing attention has been focused on improving the ability of Large Language Models (LLMs) to perform complex reasoning. Advanced methods, such as Chain-of-Thought (CoT) and its variants, are found to enhance their reasoning skills by designing suitable prompts or breaking down complex problems into more manageable sub-problems. However, little concentration has been put on exploring the reasoning process, \textit{i.e.}, we discovered that most methods resort to Direct Reasoning (DR) and disregard Indirect Reasoning (IR). This can make LLMs difficult to solve IR tasks, which are often encountered in the real world. To address this issue, we propose a Direct-Indirect Reasoning (DIR) method, which considers DR and IR as multiple parallel reasoning paths that are merged to derive the final answer. We stimulate LLMs to implement IR by crafting prompt templates incorporating the principles of contrapositive and contradiction. These templates trigger LLMs to assume the negation of the conclusion as true, combine it with the premises to deduce a conclusion, and utilize the logical equivalence of the contrapositive to enhance their comprehension of the rules used in the reasoning process. Our DIR method is simple yet effective and can be straightforwardly integrated with existing variants of CoT methods. Experimental results on four datasets related to logical reasoning and mathematic proof demonstrate that our DIR method, when combined with various baseline methods, significantly outperforms all the original methods.

cs.CL↗