arXiv subjects

Jiancan Wu

Publications and source records attributed to Jiancan Wu.

At least 19 recordsLinked to original sources

arXiv2026

PEA-DPO: Perception-Enhanced Alignment Direct Preference Optimization for MLLMs Alignment

Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences. However, its adaptation to multimodal settings remains unexplored. Through representational analysis, we identify a key limitation in multimodal preference optimization, which we term visual insensitivity: models often fail to distinguish between images and those with critical visual context removed. Our theoretical analysis further uncovers two manifestations of this problem, namely Across-Image Insensitivity and Within-Image Insensitivity. To address these challenges, we propose Perception-Enhanced Alignment DPO (PEA-DPO), a framework for multimodal LLMs alignment, which explicitly leverages visual preference signals to overcome visual insensitivity. We further provide a theoretical analysis demonstrating that PEA-DPO provably mitigates both failure modes. Empirical results demonstrate that PEA-DPO enhances sensitivity to visual context while preserving the language modeling capacity of the base model. Evaluations across three hallucination benchmarks using MLLMs of varying scales show that PEA-DPO effectively mitigates visual insensitivity, achieves stronger multimodal alignment, and substantially reduces hallucinations.

cs.CV↗

arXiv2026

ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples

Reinforcement learning (RL) has significantly enhanced the reasoning capabilities of large language models (LLMs), yet the training process remains notoriously fragile. In this work, we investigate a critical source of this instability: over-optimization, where models exploit training heuristics at the expense of generalizable reasoning. While reverse KL regularization is the standard defense against such degradation, our analysis reveals that it is often insufficient in this regime, as it fails to ensure comprehensive coverage of the reference distribution. To address this, we propose ARMOR (Anchor Rollout and Mixed Optimization for RL), a framework that shifts the paradigm from passive penalty to active sample stabilization. ARMOR comprises two key components: (1) Anchor Rollout, which leverages off-policy data from the reference policy to preserve established solution patterns; and (2) Mixed Optimization, which reformulates the policy objective to enable controlled exploration without relying on auxiliary losses. Extensive experiments on reasoning benchmarks validate that ARMOR effectively mitigates validation collapse, enabling sustained performance improvements over extended training horizons.

cs.LG↗

arXiv2026

Confidence Before Answering: A Paradigm Shift for Efficient LLM Uncertainty Estimation

Reliable deployment of large language models (LLMs) requires accurate uncertainty estimation. Existing methods are predominantly answer-first, producing confidence only after generating an answer, which measure the correctness of a specific response and limits practical usability. We study a confidence-first paradigm, where the model outputs its confidence before answering, interpreting this score as the model's probability of answering the question correctly under its current policy. We propose CoCA(Co-optimized Confidence and Answers), a GRPO reinforcement learning framework that jointly optimizes confidence calibration and answer accuracy via segmented credit assignment. By assigning separate rewards and group-relative advantages to confidence and answer segments, CoCA enables stable joint optimization and avoids reward hacking. Experiments across math, code, and factual QA benchmarks show improved calibration and uncertainty discrimination while preserving answer quality, thereby enabling a broader range of downstream applications.

cs.CL↗

arXiv2026

Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning

Large language models (LLMs) often fail to reason under temporal cutoffs: when prompted to answer from the standpoint of an earlier time, they exploit knowledge that became available only later. We study this failure through the lens of ex-ante reasoning, where a model must rely exclusively on information knowable before a cutoff. Through a systematic analysis of prompt-level interventions, we find that temporal leakage is highly sensitive to cutoff formulation and instruction placement: explicit cutoff statements outperform implicit historical framings, and prefix constraints reduce leakage more effectively than suffix constraints. These findings indicate that prompting can steer models into a temporal frame, but does not endow them with the ability to verify whether a response is temporally admissible. We further argue that supervised fine-tuning is insufficient, since ex-ante correctness is not an intrinsic property of an answer, but a relation between the answer and the cutoff. To address this gap, we propose TCFT, a Temporal Critique Fine-Tuning framework that trains models to acquire cutoff-aware temporal verification. Given a query, a cutoff, and a candidate response, TCFT teaches the model to identify post-cutoff leakage, explain temporal boundary violations, and judge temporal admissibility. Experiments with Qwen2.5-7B-Instruct and Qwen2.5-14B-Instruct show that TCFT consistently outperforms prompting and SFT baselines, reducing average leakage by 41.89 and 37.79 percentage points, respectively.

cs.AI↗

arXiv2026

Scaling Retrieval-Augmented Reasoning with Parallel Search and Explicit Merging

Deep search agents have proven effective in enhancing LLMs by retrieving external knowledge during multi-step reasoning. However, existing methods often generate a single query for retrieval at each reasoning step, limiting information coverage and introducing high noise. This may result in low signal-to-noise ratios (SNR) during search, degrading reasoning accuracy and leading to unnecessary reasoning steps. In this paper, we introduce MultiSearch, an RL-based framework that addresses these limitations through multi-query retrieval and explicit merging of retrieved information. At each reasoning step, MultiSearch generates queries from multiple perspectives and retrieves external information in parallel, expanding the scope of relevant information and mitigating the reliance on any single retrieval result. Then, the agent consolidates and refines retrieved information at the merging process, improving the SNR and ensuring more accurate reasoning. Additionally, we propose a reinforcement learning framework with a multi-process reward design to optimize agents for both multi-query retrieval and information consolidation. Extensive experiments on seven benchmarks demonstrate that MultiSearch outperforms baseline methods, enhancing the SNR of retrieval and improving reasoning performance in question-answering tasks.

cs.AI↗

arXiv2026

DynamicPO: Dynamic Preference Optimization for Recommendation

In large language model (LLM)-based recommendation systems, direct preference optimization (DPO) effectively aligns recommendations with user preferences, requiring multi-negative objective functions to leverage abundant implicit-feedback negatives and sharpen preference boundaries. However, our empirical analyses reveal a counterintuitive phenomenon, preference optimization collapse, where increasing the number of negative samples can lead to performance degradation despite a continuously decreasing training loss. We further theoretically demonstrate that this collapse arises from gradient suppression, caused by the dominance of easily discriminable negatives over boundary-critical negatives that truly define user preference boundaries. As a result, boundary-relevant signals are under-optimized, weakening the model's decision boundary. Motivated by these observations, we propose DynamicPO (Dynamic Preference Optimization), a lightweight and plug-and-play framework comprising two adaptive mechanisms: Dynamic Boundary Negative Selection, which identifies and prioritizes informative negatives near the model's decision boundary, and Dual-Margin Dynamic beta Adjustment, which calibrates optimization strength per sample according to boundary ambiguity. Extensive experiments on three public datasets show that DynamicPO effectively prevents optimization collapse and improves recommendation accuracy on multi-negative preference optimization methods, with negligible computational overhead. Our code and datasets are available at https://github.com/xingyuHuxingyu/DynamicPO.

cs.IR↗

arXiv2026

Joint Optimization of Multi-agent Memory System

Memory systems are critical for LLMs, mitigating context window limitations and supporting long-horizon user-LLM interactions. Such systems typically comprise multiple agents responsible for memory construction and retrieval. Existing approaches often optimize each agent independently under a shared global objective (e.g., downstream QA accuracy), treating other agents as a static environment. However, this design has two key limitations: (1) independent optimization ignores inter-agent dependencies and lacks agents' co-adaptation, and (2) relying solely on sparse global rewards provides limited guidance for optimizing specialized agents and causes ambiguous credit assignment. These may ultimately limit agents' effective collaboration in the memory system. To address these limitations, we propose CoMAM, a joint optimization framework that promotes collaboration among agents via end-to-end reinforcement learning and an adaptive credit assignment mechanism. Specifically, we model the multi-agent pipeline as a Markov decision process (MDP) to expose inter-agent dependencies during end-to-end training. Agents are then jointly optimized using a combination of their local task reward and an adaptively weighted global reward, enabling agents to co-adapt while receiving targeted feedback for their respective roles. Experiments show that CoMAM consistently outperforms leading memory systems, validating the effectiveness of the joint optimization framework.

cs.MA↗

arXiv2026

SepSeq: A Training-Free Framework for Long Numerical Sequence Processing in LLMs

While transformer-based Large Language Models (LLMs) theoretically support massive context windows, they suffer from severe performance degradation when processing long numerical sequences. We attribute this failure to the attention dispersion in the Softmax mechanism, which prevents the model from concentrating attention. To overcome this, we propose Separate Sequence (SepSeq), a training-free, plug-and-play framework to mitigate dispersion by strategically inserting separator tokens. Mechanistically, we demonstrate that separator tokens act as an attention sink, recalibrating attention to focus on local segments while preserving global context. Extensive evaluations on 9 widely-adopted LLMs confirm the effectiveness of our approach: SepSeq yields an average relative accuracy improvement of 35.6% across diverse domains while reducing total inference token consumption by 16.4% on average.

cs.CL↗

arXiv2026

Beyond Where to Look: Trajectory-Guided Reinforcement Learning for Multimodal RLVR

Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) for multimodal large language models (MLLMs) have mainly focused on improving final answer correctness and strengthening visual grounding. However, a critical bottleneck remains: although models can attend to relevant visual regions, they often fail to effectively incorporate visual evidence into subsequent reasoning, leading to reasoning chains that are weakly grounded in visual facts. To address this issue, we propose Trajectory-Guided Reinforcement Learning (TGRL), which guides the policy model to integrate visual evidence into fine-grained reasoning processes using expert reasoning trajectories from stronger models. We further introduce token-level reweighting and trajectory filtering to ensure stable and effective policy optimization. Extensive experiments on multiple multimodal reasoning benchmarks demonstrate that TGRL consistently improves reasoning performance and effectively bridges the gap between visual perception and logical reasoning.

cs.CV↗

arXiv2026

Bridging Perception and Reasoning: Token Reweighting for RLVR in Multimodal LLMs

Extending Reinforcement Learning with Verifiable Rewards (RLVR) to multimodal large language models (MLLMs) faces a fundamental challenge: their responses inherently interleave perception-related tokens, which ground visual content, with reasoning-related tokens, which construct reasoning chains. These token types instantiate distinct yet interdependent capacities -- visual grounding and symbolic reasoning -- making isolated optimization insufficient. Through token-level empirical analysis, we demonstrate that optimizing either perception- or reasoning-only tokens consistently underperforms full optimization, underscoring their inherent coupling. To address this, we propose a plug-and-play Token-Reweighting (ToR) strategy that explicitly models this interdependence by identifying critical tokens of both types and dynamically reweighting them during RLVR training. Applied on top of existing methods (e.g., GRPO and DAPO), ToR delivers consistent performance gains across multiple multi-modal reasoning benchmarks, achieving state-of-the-art performance with both accurate visual grounding and coherent reasoning.

cs.CV↗

arXiv2026

On the Direction of RLVR Updates for LLM Reasoning: Identification and Exploitation

Reinforcement learning with verifiable rewards (RLVR) has substantially improved the reasoning capabilities of large language models. While existing analyses identify that RLVR-induced changes are sparse, they primarily focus on the \textbf{magnitude} of these updates, largely overlooking their \textbf{direction}. In this work, we argue that the direction of updates is a more critical lens for understanding RLVR's effects, which can be captured by the signed, token-level log probability difference $Δ\log p$ between the base and final RLVR models. Through statistical analysis and token-replacement interventions, we demonstrate that $Δ\log p$ more effectively identifies sparse, yet reasoning-critical updates than magnitude-based metrics (\eg divergence or entropy). Building on this insight, we propose two practical applications: (1) a \textit{test-time extrapolation} method that amplifies the policy along the learned $Δ\log p$ direction to improve reasoning accuracy without further training; (2) a \textit{training-time reweighting} method that focuses learning on low-probability (corresponding to higher $Δ\log p$) tokens, which improves reasoning performance across models and benchmarks. Our work establishes the direction of change as a key principle for analyzing and improving RLVR.

cs.LG↗

arXiv2026

MLLMRec-R1: Incentivizing Reasoning Capability in Large Language Models for Multimodal Sequential Recommendation

Group relative policy optimization (GRPO) has become a standard post-training paradigm for improving reasoning and preference alignment in large language models (LLMs), and has recently shown strong effectiveness in LLM-based recommender systems. However, extending GRPO-based reasoning pipelines to multimodal sequential recommendation (MSR) with multimodal large language models (MLLMs) faces fundamental obstacles. First, MSR requires jointly encoding visual content for both historical interactions and multiple candidate items, causing visual tokens to dominate the input and making the cost of group-based rollout scale with history length and candidate set size, which renders GRPO-based training prohibitively expensive. Second, existing Chain-of-Thought (CoT) supervision suffers from reward inflation in recommendation scenarios, where higher training rewards do not reliably translate into improved ranking performance and may induce shortcut learning. To address these challenges, we propose MLLMRec-R1, an efficient and stable GRPO-based reasoning framework for multimodal sequential recommendation. MLLMRec-R1 textualizes visual signals offline to eliminate expensive visual tokens while preserving multimodal semantics, and constructs high-quality multimodal CoT supervision through refinement and confidence-aware assessment. Furthermore, a mixed-grained data augmentation strategy selectively injects reliable CoT samples while retaining standard training data, mitigating reward inflation and improving generalization stability. Extensive experiments on three benchmark datasets demonstrate that MLLMRec-R1 consistently outperforms state-of-the-art methods, establishing a practical and effective GRPO-based reasoning pipeline for multimodal sequential recommendation. The code is available at https://github.com/wangyu0627/MLLMRec-R1.

cs.IR↗

arXiv2026

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning

Reinforcement Learning with Verifiable Rewards (RLVR) strengthens LLM reasoning, but training often oscillates between {entropy collapse} and {entropy explosion}. We trace both hazards to the mean baseline used in value-free RL (e.g., GRPO and DAPO), which improperly penalizes negative-advantage samples under reward outliers. We propose {Quantile Advantage Estimation} (QAE), replacing the mean with a group-wise K-quantile baseline. QAE induces a response-level, two-regime gate: on hard queries (p <= 1 - K) it reinforces rare successes, while on easy queries (p > 1 - K) it targets remaining failures. Under first-order softmax updates, we prove {two-sided entropy safety}, giving lower and upper bounds on one-step entropy change that curb explosion and prevent collapse. Empirically, this minimal modification stabilizes entropy, sparsifies credit assignment (with tuned K, roughly 80% of responses receive zero advantage), and yields sustained pass@1 gains on Qwen3-8B/14B-Base across AIME 2024/2025 and AMC 2023. These results identify {baseline design} -- rather than token-level heuristics -- as the primary mechanism for scaling RLVR.

cs.LG↗

arXiv2026

Fine-grained Semantics Integration for Large Language Model-based Recommendation

Recent advances in Large Language Models (LLMs) have driven a shift in recommender systems from the discriminative paradigm to the LLM-based generative paradigm, where the recommender autoregressively generates sequences of semantic identifiers (SIDs) for target items conditioned on historical interaction. While prevalent LLM-based recommenders have demonstrated performance gains by aligning pretrained LLMs between the language space and the SID space, modeling the SID space still faces two fundamental challenges: (i) Semantically Meaningless Initialization: SID tokens are initialized by random sampling without incorporating token-specific semantics, leaving them without explicit semantic correspondence to the pretrained language space; and (ii) Coarse-grained Alignment: existing SFT-based alignment tasks primarily focus on item-level optimization while overlooking the semantics of individual tokens within SID sequences. To address these challenges, we propose TS-Rec, which can integrate Token-level Semantics into LLM-based Recommenders. Specifically, TS-Rec comprises two key components: (1) Semantic-Aware Embedding Initialization (SA-Init), which initializes SID token embeddings by applying mean pooling to the pretrained embeddings of keywords extracted by a teacher model; and (2) Token-level Semantic Alignment (TS-Align), which aligns individual tokens within the SID sequence with the shared semantics of the corresponding item clusters. Extensive experiments on three real-world benchmarks demonstrate that TS-Rec consistently outperforms traditional and generative baselines across all standard metrics. These findings confirm the effectiveness of incorporating fine-grained semantic information into LLM-based generative recommenders. The implementation is available at https://anonymous.4open.science/r/TS-Rec-Review-8F99/.

cs.IR↗

arXiv2026

MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering

Multimodal large language models (MLLMs) have demonstrated remarkable capabilities across vision-language tasks, yet their large-scale deployment raises pressing concerns about memorized private data, outdated knowledge, and harmful content. Existing unlearning approaches for MLLMs typically adapt training-based strategies such as gradient ascent or preference optimization, but these methods are computationally expensive, irreversible, and often distort retained knowledge. In this work, we propose MLLMEraser, an input-aware, training-free framework for test-time unlearning. Our approach leverages activation steering to enable dynamic knowledge erasure without parameter updates. Specifically, we construct a multimodal erasure direction by contrasting adversarially perturbed, knowledge-recall image-text pairs with knowledge-erasure counterparts, capturing both textual and visual discrepancies. To prevent unnecessary interference, we further design an input-aware steering mechanism that adaptively determines when and how the erasure direction should be applied, preserving utility on retained knowledge while enforcing forgetting on designated content. Experiments on LLaVA-1.5 and Qwen-2.5-VL demonstrate that MLLMEraser consistently outperforms state-of-the-art MLLM unlearning baselines, achieving stronger forgetting performance with lower computational cost and minimal utility degradation.

cs.LG↗

arXiv2025

LaMP-Val: Large Language Models Empower Personalized Valuation in Auction

Auctions are a vital economic mechanism used to determine the market value of goods or services through competitive bidding within a specific framework. However, much of the current research primarily focuses on the bidding algorithms used within auction mechanisms. This often neglects the potential benefits of incorporating individual users' unique preferences into the valuation process. Our theoretical and empirical analysis demonstrates that valuation errors can significantly impact the overall utility. To bridge this gap, we propose a personalized valuation framework, namely Large \underline{La}nguage \underline{M}odels-powered \underline{P}ersonalized \underline{Val}uation (LaMP-Val), which integrates Large Language Models to incorporate personalized semantic preference into users valuation process. LaMP-Val integrating three components: data, learning, and evaluation. The data component tackles the challenge of building a novel dataset specifically for LLMs fine-tuning in personalized valuation modeling. The learning component introduces a diversity template to enhance LLMs' capacity for modeling fine-grained personal valuation patterns. The evaluation component establishes a closed-loop system where LLM-generated valuations interact with bidding strategies and auction. It proposes two novel metrics to quantify valuation precision and bidding intention accuracy in personalized scenarios. Extensive experiments show that LaMP-Val more accurately captures personalized values and achieves greater profits than baseline approaches.

cs.CE↗

arXiv2025

Robust Preference Optimization via Dynamic Target Margins

The alignment of Large Language Models (LLMs) is crucial for ensuring their safety and reliability in practical applications. Direct Preference Optimization (DPO) has emerged as an efficient method that directly optimizes models using preference pairs, significantly reducing resource demands. However, the effectiveness of DPO heavily depends on the data quality, which is frequently compromised by noise. In this work, we propose $γ$-PO, a dynamic target margin preference optimization algorithm that adjust reward margins at the pairwise level. By introducing instance-specific margin calibration, $γ$-PO strategically prioritizes high-confidence pairs (those demonstrating higher reward margins) while suppressing potential noise from ambiguous pairs. Moreover, $γ$-PO is a plug-and-play method, compatible with variants of DPO that rely on reward margin between preference pairs. Across benchmarks such as AlpacaEval2 and Arena-Hard, $γ$-PO achieves an average 4.4\% improvement over other baselines, setting new benchmarks for state-of-the-art performance. Additionally, $γ$-PO requires minimal code changes and has a negligible impact on training efficiency, making it a robust solution for enhancing LLMs alignment. Our codes are available at \href{https://github.com/sunjie279/gammaPO}{https://github.com/sunjie279/gammaPO}.

cs.CL↗

arXiv2025

MiniOneRec: An Open-Source Framework for Scaling Generative Recommendation

The recent success of large language models (LLMs) has renewed interest in whether recommender systems can achieve similar scaling benefits. Conventional recommenders, dominated by massive embedding tables, tend to plateau as embedding dimensions grow. In contrast, the emerging generative paradigm replaces embeddings with compact Semantic ID (SID) sequences produced by autoregressive Transformers. Yet most industrial deployments remain proprietary, leaving two fundamental questions open: (1) Do the expected scaling laws hold on public benchmarks? (2) What is the minimal post-training recipe that enables competitive performance? We present MiniOneRec, to the best of our knowledge, the first fully open-source generative recommendation framework, which provides an end-to-end workflow spanning SID construction, supervised fine-tuning, and recommendation-oriented reinforcement learning. We generate SIDs via a Residual Quantized VAE and post-train Qwen backbones ranging from 0.5B to 7B parameters on the Amazon Review dataset. Our experiments reveal a consistent downward trend in both training and evaluation losses with increasing model size, validating the parameter efficiency of the generative approach. To further enhance performance, we propose a lightweight yet effective post-training pipeline that (1) enforces full-process SID alignment and (2) applies reinforcement learning with constrained decoding and hybrid rewards. Together, these techniques yield significant improvements in both ranking accuracy and candidate diversity.

cs.IR↗