论文检索

Deep Learning · Large Language Models

Shuo Yang, Jinda Lu, Kexin Huang, Chiyu Ma, Shaohang Wei, Yuyang Liu, Guoyin Wang, Jingren Zhou, Li Yuan

Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling reasoning capabilities of Large Language Models (LLMs). However, the sparsity of binary verifier rewards often leads to low efficiency and optimization instability. To stabilize training, existing methods typically impose token-level constraints relative to a reference policy. We identify that such constraints penalize deviations indiscriminately; this can flip verifier-determined direction when the policy attempts to outperform the reference, thereby suppressing gains. To resolve this, we propose **One-Way Policy Optimization (OWPO)**, a method based on the principle of decoupling optimization direction from update magnitude. In OWPO, the verifier dictates the update direction, while the reference policy serves only to adjust the magnitude. Specifically, OWPO applies asymmetric reweighting: it performs **Accelerated Alignment** for Inferior deviations (where the policy lags behind the reference) and **Gain Locking** for Superior deviations (where the policy surpasses the reference). Furthermore, by incorporating iterative reference updates, OWPO creates a ``Ratchet Effect'' that continuously consolidates gains. Experimental results demonstrate that OWPO outperforms strong baselines, including DAPO, OPD, and MOPD, breaking the bottleneck of fixed priors to enable continuous self-evolution without reliance on external reference models.

Probabilistic Methods · Bayesian Models and Methods

Tom Sprunck, Marcelo Pereyra, Tobías Liaudat

Modern imaging techniques heavily rely on Bayesian statistical models to address difficult image reconstruction and restoration tasks. This paper addresses the objective evaluation of such models in settings where ground truth is unavailable, with a focus on model selection and misspecification diagnosis. Existing unsupervised model evaluation methods are often unsuitable for computational imaging due to their high computational cost and incompatibility with modern image priors defined implicitly via machine learning models. We herein propose a general methodology for unsupervised model selection and misspecification detection in Bayesian imaging sciences, based on a novel combination of Bayesian cross-validation and data fission, a randomized measurement splitting technique. The approach is compatible with any Bayesian imaging sampler, including diffusion and plug-and-play samplers. We demonstrate the methodology through experiments involving various scoring rules and types of model misspecification, where we achieve excellent selection and detection accuracy with a low computational cost.

Deep Learning · Large Language Models

Jun Liu, Peilin Liu, Ruicheng Zhang, Senlei Zhang, Yanbo Chen, Ziao Wang, Jinyun Yang, mingqi wang, Shuhao Zhang, Xiaofei Liao 等

LLM applications increasingly execute as end-to-end inference pipelines that couple generation with retrieval, stateful memory, context refinement, and tool use under strict tail-latency and SLO constraints. Today, these stages are often stitched together as RPC-connected services, obscuring cross-stage queueing and interference and limiting pipeline-level compilation and resource sharing. We present SAGE (Streaming-Augmented Generative Execution), a full-stack system that treats inference pipelines as first-class compilation targets. SAGE exposes pipelines as declarative dataflows and compiles them into distributed execution plans with bounded-queue backpressure. It integrates vector search, streaming semantic state, structured memory, and refinement as operators with explicit resource/state contracts, enabling operator-level diagnosis of tail behavior. SAGE integrates pluggable generation and embedding backends and provides a unified control plane for engine management, batching, and admission under mixed workloads. On a 16-node cluster, SAGE sustains 16 requests/s at $>700$ tokens/request with 1 ms median scheduling overhead, and achieves near-linear scale-out to 16 nodes (11.4$\times$ throughput at 16 nodes), and reduces p99 latency by 57\% under multi-pipeline contention versus simultaneous admission.

Deep Learning · Large Language Models

Beidi Zhao, wenlong deng, Xinting Liao, Yushu Li, Nazim Shaikh, Yao Nie, Xiaoxiao Li

While Retrieval-Augmented Generation (RAG) is one of the dominant paradigms for enhancing Large Vision-Language Models (LVLMs) on knowledge-based VQA tasks, recent work attributes RAG failures to insufficient attention towards the retrieved context, proposing to reduce the attention allocated to image tokens. In this work, we identify a distinct failure mode that previous study overlooked: Attention Distraction (AD). When the retrieved context is sufficient (highly relevant or including the correct answer), the retrieved text suppresses the visual attention globally, and the attention on image tokens shifts away from question-relevant regions. This leads to failures on questions the model could originally answer correctly without the retrieved text. To mitigate this issue, we propose MAD-RAG, a training-free intervention that decouples visual grounding from context integration through a dual-question formulation, combined with attention mixing to preserve image-conditioned evidence. Extensive experiments on OK-VQA, E-VQA, and InfoSeek demonstrate that MAD-RAG consistently outperforms existing baselines across different model families, yielding absolute gains of up to 4.76%, 9.20%, and 6.18% over the vanilla RAG baseline. Notably, MAD-RAG rectifies up to 74.68% of failure cases with negligible computational overhead.

Deep Learning · Large Language Models

Skyler Seto, Pierre Ablin, Anastasiia Filippova, Jiayuan Ye, Louis Béthune, Angelos Katharopoulos, David Grangier

Language models achieve impressive performance on a variety of knowledge, language, and reasoning tasks due to the scale and diversity of pretraining data available. The standard training recipe is a two-stage paradigm: pretraining first on the full corpus of data followed by specialization on a much smaller subset of high quality, specialized data from the full corpus. In the multi-domain setting, this involves continued pretraining of multiple models on each specialized domain, referred to as split model training. We propose a method for pretraining multiple models independently over a general pretraining corpus, and determining the optimal compute allocation between pretraining and continued pretraining using scaling laws. Our approach accurately predicts the loss of a model of size $N$ with $D$ pretraining and $D'$ specialization tokens, and extrapolates to larger model sizes and number of tokens. Applied to language model training, our approach improves performance consistently across common sense knowledge and reasoning benchmarks across different model sizes and compute budgets.

Deep Learning · Large Language Models

Ke Yang, Zixi Chen, Xuan He, Jize Jiang, Michel Galley, Chenglong Wang, Jianfeng Gao, Jiawei Han, Chengxiang Zhai

Long-term memory is essential for large language model (LLM) agents operating in complex environments, yet existing memory designs are either task-specific and non-transferable, or task-agnostic but less effective due to low task-relevance and context explosion from raw memory retrieval. We propose PlugMem, a task-agnostic plugin memory module that can be attached to arbitrary LLM agents without task-specific redesign. Motivated by the fact that decision-relevant information is concentrated as abstract knowledge rather than raw experience, we draw on cognitive science to structure episodic memories into a compact, extensible knowledge-centric memory graph that explicitly represents propositional and prescriptive knowledge. This representation enables efficient memory retrieval and reasoning over task-relevant knowledge, rather than verbose raw trajectories, and departs from other graph-based methods like GraphRAG by treating knowledge as the unit of memory access and organization instead of entities or text chunks. We evaluate PlugMem unchanged across three heterogeneous benchmarks (long-horizon conversational question answering, multi-hop knowledge retrieval, and web agent tasks). The results show that PlugMem consistently outperforms task-agnostic baselines and exceeds task-specific memory designs, while also achieving the highest information density under a unified information-theoretic analysis.

Deep Learning · Large Language Models

Zeyu Zhang, Guohao Li, Zhenchang Xing, Alexandros Apostolopoulos, Yu Lin Lee, Liang Zheng

The ability to use tools is fundamental for large language model (LLM) agents. Given a task, existing systems use LLMs to plan and generate tool calls, which are executed by real-world tools to complete the task. However, tool calls are prone to errors because they are derived merely from LLM intrinsic capabilities. What is more, while it is useful to let LLMs iteratively refine the tool-call sequence using execution results from real tools, this process can be expensive and lead to unsafe results. To improve LLM tool calls and address issues caused by using real tools for refinement, we introduce Gecko, a comprehensive environment that simulates tool responses using a combination of rules and LLMs. Specifically, Gecko checks the validity of tool calls including input arguments and tool names, synthesizes reasonable responses that adhere to the output schema, and assesses whether all task objectives have been achieved. These three types of feedback provided by Gecko allow LLMs to refine their tool calls, forming a simple yet effective test-time scaling method named GATS. On BFCLv3 and $\tau^2$-bench, GATS consistently improves the tool calling performance of various LLMs including GPT-4o, GPT-5, and Gemini-3.0-pro. We further discuss working mechanisms of our method and share future possibilities.

Deep Learning · Large Language Models

Xinxin You, Xien Liu, Chenwei Yan, Siqi Song, Chen Ning, Kaiyin Zhou, shaohui liu, Ji Wu

Large Language Models (LLMs) frequently generate output that contradicts explicit input evidence, limiting their reliability in real-world applications. We identify cognitive inertia in LLMs—a tendency to overly rely on co-occurrence associations learned during pretraining and to resist adaptation when conflicting input evidence appears—as a critical factor behind such hallucinations. We further empirically show that adherence to input evidence declines as co-occurrence associations are strengthened—driven by either higher data frequency or intensified training. Inspired by human counter-inertial thinking, we propose an adaptive counter-inertial reasoning framework that probes input-related cognitive inertia in the LLM and generates adaptive counter-inertial reminders, which are then injected into the prompt to promote evidence-based reasoning. Experiments on co-occurrence induction datasets show that LLMInertia reduces hallucination rates by up to 35\% and improves accuracy by up to 35.68\%. Extensive evaluations on four context-rich summarization and QA datasets, across three LLM backbones of varying scales, further validate its effectiveness and robustness. Our work provides new insight into the causes of input-unfaithful hallucinations in LLMs, contributing to the development of more reliable AI.

Deep Learning · Large Language Models

Sora Miyamoto, Daisuke Oba, Naoaki Okazaki

Tree-search decoding is an effective form of test-time scaling for large language models (LLMs), but real-world deployment imposes a fixed per-query token budget that varies across settings. Existing tree-search policies are largely budget-agnostic, treating the budget as a termination condition, which can lead to late-stage over-branching or premature termination. We propose Budget-Guided MCTS (BG-MCTS), a tree-search decoding algorithm that aligns its search policy with the remaining token budget: it starts with broad exploration, then prioritizes refinement and answer completion as the budget depletes while reducing late-stage branching from shallow nodes. BG-MCTS consistently outperforms budget-agnostic tree-search baselines across different budgets on MATH500 and AIME24/25 with open-weight LLMs.

Deep Learning · Large Language Models

Nihal Nayak, Paula Rodriguez-Diaz, Neha Hulkund, Sara Beery, David Alvarez-Melis

Instruction fine-tuning of large language models (LLMs) often involves selecting a subset of instruction training data from a large candidate pool, using a small query set from the target task. Despite growing interest, the literature on targeted instruction selection remains fragmented and opaque: methods vary widely in selection budgets, often omit zero-shot baselines, and frequently entangle the contributions of key components. As a result, practitioners lack actionable guidance on selecting instructions for their target tasks. In this work, we aim to bring clarity to this landscape by disentangling and systematically analyzing the two core ingredients: data representation and selection algorithms. Our framework enables controlled comparisons across models, tasks, and budgets. We find that only gradient-based data representations choose subsets whose similarity to the query consistently predicts performance across datasets and models. While no single method dominates, gradient-based representations paired with a greedy round-robin selection algorithm tend to perform best on average at low budgets, but these benefits diminish at larger budgets. Finally, we unify several existing selection algorithms as forms of approximate distance minimization between the selected subset and the query set, and support this view with new generalization bounds. More broadly, our findings provide critical insights and a foundation for more principled data selection in LLM fine-tuning.

Deep Learning · Large Language Models

Son Nguyen, Xinyuan Liu, Ransalu Senanayake

Users increasingly face the challenge of selecting an appropriate LLM for a given task from a rapidly growing pool of LLMs, each with distinct but often opaque latent properties. Compounding this challenge, users may lack the vocabulary or awareness to explicitly articulate the characteristics they value in an LLM's responses or deployment. We propose an interaction-efficient active learning framework in which a dueling bandit algorithm iteratively selects pairs of LLMs, collects user feedback about their responses, and updates its belief about the user's latent preferences. We introduce a novel belief-aware upper confidence bound strategy that balances exploration of the model pool with exploitation of inferred preferences, enabling efficient alignment between user needs and LLM capabilities under user-specified cost and time budgets. Through diverse experiments on LLMs and human studies, we experimentally verify that our model can efficiently match users to LLMs at a lower cost.

Deep Learning · Large Language Models

Zhiyuan Han, Beier Zhu, Wenwen Tong, Pengyang Shao, Peipei Song, Xinyi Wang, Jiangnan Chen, Lewei Lu, Xun Yang

Recent Omni-MLLMs are driving a paradigm shift in multimodal emotion recognition from label-only prediction toward *Multimodal Emotion Reasoning* (MER), where models output both emotions and textual explanations grounded in visual, acoustic, and linguistic signals. However, we show that current emotion-oriented Omni-MLLMs still lack *reliable omni-modal perception*: they (i) underutilize multimodal cues in their reasoning trajectories and (ii) exhibit unfaithful behavior, often hallucinating modality-specific statements from other modalities. Building on these insights, we propose **OPPO** (**O**mni-**P**erception **P**olicy **O**ptimization), a reinforcement learning framework that explicitly optimizes multimodal perception. First, an Omni-Perception Reward decomposes ground-truth reasoning into fine-grained visual, acoustic, and emotion cues and rewards trajectories that semantically recover these cues. Second, an Omni-Perception Loss compares the policy under full and unimodally masked inputs, applying a KL penalty only to modality-specific evidence tokens to suppress cross-modal hallucination. We further introduce *MEP-Bench*, a diagnostic benchmark that quantifies *utilization* and *faithfulness*. Experiments show that OPPO achieves state-of-the-art performance on MER-UniBench and substantially improves utilization and faithfulness scores on MEP-Bench, highlighting the importance of sufficient and faithful omni perception for multimodal emotion reasoning. The code is provided in the Supplementary Materials.

Deep Learning · Large Language Models

Jakub Binkowski, Kamil Adamczewski, Tomasz Kajdanowicz

Large language models frequently exhibit hallucinations: fluent and confident outputs that are factually incorrect or unsupported by the input context. While recent hallucination detection methods have explored various features derived from attention maps, the underlying mechanisms they exploit remain poorly understood. In this work, we propose SinkProbe, a hallucination detection method grounded in the observation that hallucinations are deeply entangled with attention sinks - tokens that accumulate disproportionate attention mass during generation - indicating a transition from distributed, input-grounded attention to compressed, prior-dominated computation. Importantly, although sink scores are computed solely from attention maps, we find that the classifier preferentially relies on sinks whose associated value vectors have large norms. Moreover, we show that previous methods implicitly depend on attention sinks by establishing their mathematical relationship to sink scores. Our findings yield a novel hallucination detection method grounded in theory that produces state-of-the-art results across popular datasets and LLMs.

Reinforcement Learning · Everything Else

Xinyu Zhang

Asynchronous reinforcement learning enables high-throughput training but introduces *policy lag*, where experiences are collected under stale policy weights. We identify a key phenomenon in code generation: **gradient variance scales exponentially with task difficulty under staleness**, because hard tasks have narrow solution spaces corresponding to sharp loss landscape curvature (high Hessian eigenvalues). We formalize this as a *staleness budget optimization problem* and prove that the optimal allocation follows an exponential decay: $\eta^*(d) = \eta_{base} \cdot e^{-\lambda d}$ where $\lambda = \alpha/2$ is half the Hessian growth rate. Building on this theory, we propose ACEAS (**A**daptive **C**urriculum with **E**xecution-**A**ware **A**sync **S**cheduling), combining bandit-based curriculum selection, execution-aware staleness budgets, and curriculum-staleness coupling derived from first principles. Our mechanistic analysis validates the theoretical predictions: the "safe zone" of gradient coherence follows the derived exponential boundary. On code generation benchmarks, ACEAS achieves over 2$\times$ higher throughput than synchronous baselines while improving Pass@1 from 39.7% to 60.1%, demonstrating that principled staleness control grounded in loss landscape geometry enables efficient asynchronous curriculum learning.

Deep Learning · Large Language Models

Lijinghua Zhang, Michelle Bang, Hengrui Cai

This position paper argues that *prompting intent should be audited in LLM-assisted peer review*, moving beyond the sole detection or disclosure of LLM usage. As major conferences increasingly allow LLM assistance and deploy mechanisms for detecting LLM-generated text, a critical gap remains: usage alone does not determine risk. A more consequential variable is *prompting intent*, the objective or stance encoded in how an LLM is instructed, which can systematically shape review framing and tone. We advocate an *intent-centric auditing perspective* that treats prompting intent as latent and infers relevant signals from the review text. Because intent is unobservable in real deployments, we train an intent detector using synthetically labeled LLM-generated reviews. Among ICLR 2026 reviews previously flagged for substantial LLM usage, we apply our detector to infer prompting intent and find coherent linguistic and structural signatures associated with directional prompting, along with systematic associations with review ratings, confidence, and paper acceptance decisions. We conclude with practical considerations for auditing LLM-assisted peer review, with an emphasis on procedural transparency and human-in-the-loop oversight.

Deep Learning · Large Language Models

Cagatay Yildiz, Nishaanth Kanna, Nitin Sharma, Matthias Bethge, Beyza Ermis

Continual learning (CL) in large language models (LLMs) is an evolving domain that focuses on developing efficient and sustainable training strategies to adapt models to emerging knowledge and achieve robustness in dynamic environments. Our primary emphasis is on continual domain-adaptive pretraining, a process designed to equip LLMs with the ability to integrate new information from various domains while retaining previously learned knowledge. Since existing works concentrate mostly on continual fine-tuning for a limited selection of downstream tasks or training domains, we introduce a new benchmark designed to measure the adaptability of LLMs to changing pretraining data landscapes. We further examine the impact of model size on learning efficacy and forgetting, as well as how the progression and similarity of emerging domains affect the knowledge transfer within these models. Our findings uncover several key insights: (i) continual pretraining consistently improves <1.5B models studied in this work and is also superior to domain adaptation, (ii) larger models always achieve better perplexity than smaller ones when continually pretrained on the same corpus, (iii) smaller models are particularly sensitive to continual pretraining, showing the most significant rates of both learning and forgetting, (iv) continual pretraining boosts downstream task performance of GPT-2 family, (v) continual pretraining enables LLMs to specialize better when the sequence of domains shows semantic similarity while randomizing training domains leads to better transfer and final performance otherwise. We posit that our research establishes a new benchmark for CL in LLMs, providing a more realistic evaluation of knowledge retention and transfer across diverse domains.

Vahid Shahverdi, Giovanni Luca Marchetti, Georg Bökman, Kathlén Kohn

We investigate the relation between end-to-end equivariance and layerwise equivariance in deep neural networks. We prove the following: For a network whose end-to-end function is equivariant with respect to group actions on the input and output spaces, there is a parameter choice yielding the same end-to-end function such that its layers are equivariant with respect to some group actions on the latent spaces. Our result assumes that the parameters of the model are identifiable in an appropriate sense. This identifiability property has been established in the literature for a large class of networks, to which our results apply immediately, while it is conjectural for others. The theory we develop is grounded in an abstract formalism, and is therefore architecture-agnostic. Overall, our results provide a mathematical explanation for the emergence of equivariant structures in the weights of neural networks during training -- a phenomenon that is consistently observed in practice.

Deep Learning · Graph Neural Networks

Daria Fomina, Daniil Krasylnikov, Alexey Boykov, Andrey Dolgovyazov, Vyacheslav Zhdanovskiy, Fedor Velikonivtsev

Graph Neural Networks (GNNs) are bottlenecked by sparse, irregular memory access. Popular frameworks such as DGL and PyTorch Geometric support general message passing, but complex layers often materialize edge-wise intermediates, increasing memory traffic and limiting scalability on large graphs. We take an I/O- and arithmetic-intensity--centric view and show that widely used layers fall into three kernel families: SpMM-based convolutions, reduction-based aggregations, and attention-based layers (GATv2/Graph Transformer). For each family, we develop GPU kernels that reduce data movement, improve locality, and remain robust across realistic graphs. We also study graph reordering and find that its impact depends on the kernel mapping: it benefits neighbor-parallel (gather-dominated) kernels more consistently than feature-parallel designs. Empirically, our fused attention kernels reach up to **3.9**$\times$ speedup for Graph Transformer (median **1.6**$\times$), with Tensor Core (block-sparse) variants up to **7.3**$\times$ on locally dense graphs; for GATv2 we reach up to **8.5**$\times$ speedup (median **2.0**$\times$) while reducing peak memory by up to **76**$\times$ (median **6**$\times$). Our degree-aware reduction kernels achieve up to **10**$\times$ speedup (median **2.6**$\times$). For SpMM-based layers, properly cached cuSPARSE achieves up to **8**$\times$ speedup over DGL and outperforms evaluated custom baselines in the majority of evaluations. We release our implementations as drop-in replacements to support reproducible, hardware-aware GNN acceleration.

Deep Learning · Graph Neural Networks

Simon Forbat, Rainer Gemulla

Mixup is a widely used data augmentation technique that constructs new training examples by interpolating between existing ones. While simple and effective in domains like vision and language, applying mixup to graph data is non-trivial and there is no independent empirical evidence for its effectiveness. To fill this gap, we conducted an extensive evaluation study following a unified, established evaluation protocol for graph classification. In contrast to prior results, we found that none of the state-of-the-art mixup methods yielded statistically significant improvements over the no-mixup baseline. To obtain further insights, we analyzed the graphs generated from these mixup methods from an interpolation perspective. We found that (i) many mixup methods failed to interpolate well, (ii) high interpolation error led to performance degradation, and (iii) even good interpolation properties did not lead to performance improvements. Our findings question the efficacy of existing graph mixup methods and highlight the need for a more rigorous exploration and evaluation.

Deep Learning · Graph Neural Networks

Solveig Wittig, Antonis Vasileiou, Robert R. Nerem, Timo Stoll, Floris Geerts, Yusu Wang, Christopher Morris

In recent years, there has been growing interest in understanding neural architectures' ability to learn to execute discrete algorithms, a line of work often referred to as neural algorithmic reasoning. The goal is to integrate algorithmic reasoning capabilities into larger neural pipelines. Many such architectures are based on (message-passing) graph neural networks (MPNNs), owing to their permutation equivariance and ability to deal with sparsity and variable-sized inputs. However, existing work is either largely empirical and lacks formal guarantees or it focuses solely on expressivity, leaving open the question of when and how such architectures generalize beyond a finite training set. In this work, we propose a general theoretical framework that characterizes the necessary conditions under which MPNNs can learn an algorithm from a training set of small instances and provably approximate its behavior on inputs of arbitrary size. Our framework applies to a broad class of algorithms, including single-source shortest paths, minimum spanning trees, and general dynamic programming problems, such as the $0$-$1$ knapsack problem. In addition, we establish impossibility results for a wide range of algorithmic tasks, showing that standard MPNN cannot compute them. We derive more expressive MPNN-like architectures that overcome these limitations. Finally, we refine our analysis for the Bellman–Ford algorithm, yielding substantially smaller required training sets and significantly extending the recent work of Nerem et al., 2025 by allowing for a differentiable regularization loss. Empirical results largely support our theoretical findings.