🎯 What it does: Analyze the low-frequency bias in lightweight networks from a frequency perspective, and design GmNet by introducing a simple GLU to enhance high-frequency information.
🎯 What it does: Proposed the GneissWeb dataset, with a scale of approximately 10 trillion words, aiming to meet the dual demands of scale and quality for LLM Stage-1 pre-training;
GNN Explanations that do not Explain and How to find Them
Steve Azzolin (University of Trento), Sagar Malhotra (TU Wien)
CodeExplainability and InterpretabilityGraph Neural NetworkImageTextGraphBenchmark
🎯 What it does: This paper studies self-explaining graph neural networks (SE-GNN), revealing that they may generate 'degenerate' explanations unrelated to actual predictions, and proposes attack methods, evaluation benchmarks, and more reliable credibility metrics.
CodeClassificationRepresentation LearningGraph Neural NetworkTransformerLarge Language ModelSupervised Fine-TuningTextGraph
🎯 What it does: Propose a framework that uses graph neural networks (GNNs) as judges for pseudo-label learning and fine-tuning of large language models (LLMs), specifically designed for semi-supervised node classification under text attribute graphs (TAG) with scarce labels.
Go Beyond Earth: Understanding Human Actions and Scenes in Microgravity Environments
Di Wen (Karlsruhe Institute of Technology), Rainer Stiefelhagen (Karlsruhe Institute of Technology)
CodeRecognitionGenerationRetrievalConvolutional Neural NetworkTransformerSupervised Fine-TuningVideoTextMultimodalityBenchmarkPhysics Related
🎯 What it does: This paper constructs and releases the MicroG-4M dataset and MicroG-Bench benchmark, providing 4,759 segments of 3-second microgravity videos with fine-grained action annotations, 1,280 high-quality captions, and 7,428 visual question-answer pairs, supporting action recognition, video caption generation, and visual question answering tasks in microgravity environments.
Go-Browse: Training Web Agents with Structured Exploration
Apurva Gandhi (Carnegie Mellon University), Graham Neubig (Carnegie Mellon University)
CodeTransformerLarge Language ModelSupervised Fine-TuningAgentic AITextSequential
🎯 What it does: Automatically collect high-quality web proxy data by structuring exploration of the web environment, and use this data to fine-tune a 7B LLM to improve task success rates on WebArena.
Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems
Zherui Li (Beijing University of Posts and Telecommunications), Junfeng Fang (National University of Singapore)
CodeAnomaly DetectionAdversarial AttackTransformerLarge Language ModelAgentic AITextChain-of-Thought
🎯 What it does: Investigate and evaluate misinformation injection attacks in multi-agent systems (MAS), propose a novel red-team dataset MISINFOTASK, and design a two-phase training-free defense framework ARGUS
GoalRank: Group-Relative Optimization for a Large Ranking Model
Kaike Zhang (Kuaishou Technology), Kun Gai (Kuaishou Technology)
CodeRecommendation SystemReinforcement Learning from Human FeedbackTextSequential
🎯 What it does: Proposed a large-scale single-stage generative ranking model called GoalRank, which directly generates recommendation lists using a large generator, replacing the traditional multi-stage generate-evaluate architecture.
Goedel-Prover-V2: Scaling Formal Theorem Proving with Scaffolded Data Synthesis and Self-Correction
Yong Lin (Princeton University), Chi Jin (Princeton University)
CodeAI Code AssistantTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningTextChain-of-Thought
🎯 What it does: This paper proposes and releases the open-source automated theorem proving model series Goedel-Prover-V2 (in two sizes: 8B and 32B), which can generate complete proofs using the Lean 4 compiler and supports self-correction;
🎯 What it does: Proposed a flow matching strategy distillation framework named GS-flow, which leverages Q-guided priors and entropy regularization to enhance the inference speed, accuracy, and explorability of sparse multi-modal strategies.
🎯 What it does: Proposed a generative sequence generation framework named GoR, transforming ordinal regression tasks into autoregressive token sequence prediction problems.
GoT-R1: Unleashing Reasoning Capability of Autoregressive Visual Generation with Reinforcement Learning
Chengqi Duan (HKU MMLAB), Xihui Liu (HKU MMLAB)
CodeGenerationTransformerLarge Language ModelReinforcement LearningVision Language ModelImageTextMultimodalityChain-of-Thought
🎯 What it does: Propose the GoT-R1 framework, which enhances the semantic-space reasoning capabilities of autoregressive visual generation models through reinforcement learning, enabling the model to self-discover more efficient reasoning strategies.
GPG: A Simple and Strong Reinforcement Learning Baseline for Model Reasoning
Xiangxiang Chu (AMAP, Alibaba Group), Yong Wang (AMAP, Alibaba Group)
CodeLarge Language ModelReinforcement LearningTextMultimodalityBenchmark
🎯 What it does: Propose a simplified reinforcement learning (RL) method based on Group Policy Gradient (GPG) to enhance the reasoning capabilities of large language models and multimodal models.
GPS: Graph-guided Proactive Information Seeking in Large Language Models
Ruiqing Li (Peking University), Xu Chu (Peking University)
CodeRetrievalComputational EfficiencyReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningTextRetrieval-Augmented Generation
🎯 What it does: Propose the GPS framework, enabling LLMs to actively ask clarifying questions in RAG scenarios. First, construct a conditional reasoning DAG and dynamically prune the graph based on user feedback during traversal, ultimately achieving precise answers.
GPTailor: Large Language Model Pruning Through Layer Cutting and Stitching
Guinan Su (Max Planck Institute for Intelligent Systems), Jonas Geiping (Max Planck Institute for Intelligent Systems)
CodeCompressionComputational EfficiencyTransformerLarge Language ModelTextBenchmark
🎯 What it does: Propose a zeroth-order optimization method (GPTAILOR) that achieves efficient compression of large language models by pruning, selecting, and fusing layers across multiple fine-tuned models.
GRACE: Generative Representation Learning via Contrastive Policy Optimization
Jiashuo Sun (University of Illinois Urbana Champaign), Jiawei Han (University of Illinois Urbana Champaign)
CodeExplainability and InterpretabilityRepresentation LearningTransformerLarge Language ModelReinforcement LearningContrastive LearningText
🎯 What it does: Propose the GRACE framework, which rewrites contrastive learning signals as rewards to guide LLMs in generating interpretable reasoning texts (rationales) and encodes them into high-quality embeddings, thereby transforming LLMs from 'black-box encoders' into interpretable representation learners.
🎯 What it does: Propose a dynamic density control framework GDAGS based on Gradient Direction Consistency (GCR) to improve splitting and cloning operations in 3D Gaussian Splatting, thereby enhancing image quality for novel view synthesis and reducing the number of Gaussians.
Gradient-Normalized Smoothness for Optimization with Approximate Hessians
Andrei Semenov (EPFL), Nikita Doikov (Cornell University)
CodeOptimizationBenchmark
🎯 What it does: This paper proposes a new optimization framework called Gradient-Normalized Smoothness, and based on this concept, designs a gradient regularized Newton algorithm using approximate Hessian, which can achieve global convergence on both convex and non-convex problems;
🎯 What it does: This paper proposes a method called GradFix, which transfers task vectors from an old pre-trained model to a new model through gradient sign masking; this method only requires a small number of labeled samples to complete the transfer without additional fine-tuning.
GradPruner: Gradient-guided Layer Pruning Enabling Efficient Fine-Tuning and Inference for LLMs
Wei Huang (Ant Group), Yinggui Wang (Ant Group)
CodeComputational EfficiencyTransformerLarge Language ModelSupervised Fine-TuningTextBiomedical Data
🎯 What it does: Propose GradPruner, an efficient LLM fine-tuning and inference method that evaluates parameter importance using early fine-tuning gradients and performs hierarchical pruning and merging.
🎯 What it does: This paper proposes the GRAPHITE framework, which introduces feature nodes corresponding to discrete features, connecting nodes with similar features indirectly through feature edges, thereby significantly enhancing graph homophily and improving node classification performance on heterogeneous graphs without altering the model architecture.
Graph Random Features for Scalable Gaussian Processes
Matthew Zhang (University of Cambridge), Isaac Reid (University of Cambridge)
CodeOptimizationComputational EfficiencyGraph
🎯 What it does: Developed a framework that leverages Graph Random Features (GRF) to enable scalable Gaussian processes for Bayesian inference and optimization on large-scale graph nodes.
GraphOmni: A Comprehensive and Extensible Benchmark Framework for Large Language Models on Graph-theoretic Tasks
Hao Xu (Chinese University of Hong Kong), Tianshu Yu (Chinese University of Hong Kong)
CodeLarge Language ModelReinforcement LearningPrompt EngineeringTextGraphBenchmarkChain-of-Thought
🎯 What it does: This paper constructs a multidimensional graph theory task benchmark named GRAPHOMNI, designed to systematically evaluate the reasoning capabilities of large language models across different graph types, serialization formats, and prompting schemes.
GraphPlanner: Graph Memory-Augmented Agentic Routing for Multi-Agent LLMs
Tao Feng (University of Illinois Urbana Champaign), Jiaxuan You (University of Illinois Urbana Champaign)
CodeOptimizationComputational EfficiencyGraph Neural NetworkLarge Language ModelReinforcement LearningAgentic AIText
🎯 What it does: This paper studies the routing problem in multi-agent large language models, proposing a heterogeneous graph memory-enhanced agent router called GraphPlanner, which can automatically generate multi-step workflows for each query and execute LLMs with different roles;
🎯 What it does: This paper proposes and implements GraphUniverse, a framework capable of generating a family of graphs with semantic consistency and controllable structural properties, for systematically evaluating the inductive generalization of graph learning models on unseen graphs;
Grasp Any Region: Towards Precise, Contextual Pixel Understanding for Multimodal LLMs
Haochen Wang (Chinese Academy of Sciences), Zhaoxiang Zhang (Chinese Academy of Sciences)
CodeSegmentationTransformerLarge Language ModelVision Language ModelImageVideoTextMultimodalityBenchmark
🎯 What it does: Proposes the GAR model, which can precisely understand arbitrary mask regions in a single image and perform multi-region interaction reasoning.
🎯 What it does: Constructed the Magneton environment, including a large-scale protein substructure dataset, training framework, and evaluation benchmark, and proposed a substructure-tuning method that uses annotated substructure information to perform supervised fine-tuning on pre-trained protein encoders.
GRL-SNAM: Geometric Reinforcement Learning with Differential Hamiltonians for Navigation and Mapping in Unknown Environments
Aditya Sai Ellendula (University of Texas at Austin), Chandrajit L. Bajaj (University of Texas at Austin)
CodeAutonomous DrivingOptimizationReinforcement LearningSimultaneous Localization and Mapping
🎯 What it does: Propose the GRL-SNAM framework, treating simultaneous navigation and mapping as Hamiltonian dynamics under local perception, using the gradient of the energy field to directly generate control actions;
🎯 What it does: Propose a theoretical framework defining 'informativeness' and 'utility' in dataset distillation, and build the InfoUtil method by combining maximization of informativeness and utility to generate high-quality distilled datasets.
Grounding Generative Planners in Verifiable Logic: A Hybrid Architecture for Trustworthy Embodied AI
Feiyu Wu (Xidian University), HUI LI
CodeSafty and PrivacyRobotic IntelligenceTransformerLarge Language ModelVision Language ModelTextMultimodalityBenchmark
🎯 What it does: Proposed the Verifiable Iterative Refinement Framework (VIRF), a neuro-symbolic architecture that couples a large language model (LM) planner with a Logic Tutor based on OWL 2 logic, for achieving safe and verifiable robot planning in home environments.
CodeTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelMultimodalityBenchmark
🎯 What it does: Proposes a new task paradigm called 'Grounding-IQA', enabling fine-grained image quality assessment in multimodal language models, including detailed description with key region localization (GIQA-DES) and region-based quality visual question answering (GIQA-VQA).
Group Critical-token Policy Optimization for Autoregressive Image Generation
Guohui Zhang (University of Science and Technology of China), Feng Zhao (University of Science and Technology of China)
CodeGenerationOptimizationReinforcement Learning from Human FeedbackTransformerImageMultimodalityBenchmark
🎯 What it does: Developed a Group Critical-token Policy Optimization (GCPO) method that optimizes only critical tokens in autoregressive image generation to improve RLVR training effectiveness.
CodeRepresentation LearningTransformerLarge Language ModelText
🎯 What it does: Proposed a unified position encoding framework based on group actions, GRAPE, which includes two major categories: multiplicative (SO(d) rotations) and additive (GL near-identity matrices).
Group-Relative REINFORCE Is Secretly an Off-Policy Algorithm: Demystifying Some Myths About GRPO and Its Friends
Chaorui Yao (University of California, Los Angeles), Bolin Ding (Alibaba Group)
CodeReinforcement LearningText
🎯 What it does: This paper provides a theoretical analysis of Group-Relative REINFORCE, revealing for the first time that its essence is an off-policy algorithm. Based on this, it proposes two general improvement principles: regularization updates and data distribution adjustment. Furthermore, it unifies and reinterprets algorithms such as GRPO, OPMD, and AsymRE, and introduces data-weighting strategies such as RED-DROP and RED-WEIGHT.
🎯 What it does: Proposes the GT-Space framework, achieving heterogeneous collaborative perception across different sensing modalities; by constructing a shared feature space based on real labels, simplifying the feature alignment and fusion process.
GTR-Bench: Evaluating Geo-Temporal Reasoning in Vision-Language Models
Qinghongbing Xie (Tsinghua University), Long ZENG
CodeLarge Language ModelVision Language ModelImageVideoMultimodalityBenchmark
🎯 What it does: Proposed and implemented the GTR-Bench benchmark to evaluate the ability of vision-language models to perform geospatial-temporal reasoning by integrating maps and videos in large-scale multi-camera networks.
🎯 What it does: Proposed the K-step GUI Transition self-supervised inverse dynamics task and developed the GUI-Shift RL framework to train VLM agents using unlabeled GUI trajectories.
Guided Speculative Inference for Efficient Test-Time Alignment of LLMs
Jonathan Geuter (Harvard SEAS Kempner Institute), David Alvarez-Melis (Harvard SEAS Kempner Institute)
CodeComputational EfficiencyTransformerLarge Language ModelReinforcement LearningText
🎯 What it does: Propose the Guided Speculative Inference (GSI) method, which performs reward-guided best-N sampling during inference for large language models and combines speculative decoding for acceleration.
GuidedBench: Measuring and Mitigating the Evaluation Discrepancies of In-the-wild LLM Jailbreak Methods
Ruixuan Huang (Hong Kong University of Science and Technology), Shuai Wang (Hong Kong University of Science and Technology)
CodeSafty and PrivacyAdversarial AttackTransformerLarge Language ModelTextBenchmark
🎯 What it does: Propose GuidedBench, which includes a fine-grained dataset for LLM malicious question answering and a case-guided evaluation system called GuidedEval, to assess the effectiveness of jailbreak methods.
GuidedSampling: Steering LLMs Towards Diverse Candidate Solutions at Inference-Time
Divij Handa (Arizona State University), Chitta Baral (Arizona State University)
CodeGenerationOptimizationTransformerLarge Language ModelSupervised Fine-TuningTextChain-of-Thought
🎯 What it does: Propose a new algorithm called GUIDEDSAMPLING, which first generates diverse concepts or theorems during the exploration phase, and then uses these concepts to generate candidate solutions during the generation phase, thereby improving the diversity and accuracy of solutions.
h-MINT: Modeling Pocket-Ligand Binding with Hierarchical Molecular Interaction Network
Yanru Qu (University of Illinois Urbana-Champaign), Ge Liu (University of Illinois Urbana-Champaign)
CodeDrug DiscoveryGraph Neural NetworkTransformerBiomedical Data
🎯 What it does: Developed a protein-ligand interaction modeling framework based on overlapping BPE tokenization and hierarchical molecular interaction networks (h-MINT), significantly improving binding affinity prediction and virtual screening performance.
HackWorld: Evaluating Computer-Use Agents on Exploiting Web Application Vulnerabilities
Xiaoxue Ren (Zhejiang University), Terry Yue Zhuo (Monash University)
CodeSafty and PrivacyLarge Language ModelAgentic AITextMultimodalityBenchmark
🎯 What it does: Proposed the HackWorld framework, systematically evaluated the ability of computer-using agents (CUA) to identify and exploit real Web application vulnerabilities through visual interaction, and conducted experiments in 36 CTF environments containing real vulnerabilities.
Half-order Fine-Tuning for Diffusion Model: A Recursive Likelihood Ratio Optimizer
Tao Ren (Peking University), Yijie Peng (Peking University)
CodeGenerationReinforcement Learning from Human FeedbackSupervised Fine-TuningReinforcement LearningPrompt EngineeringDiffusion modelImageVideoTextChain-of-Thought
🎯 What it does: Proposed a half-order gradient estimator called Recursive Likelihood Ratio (RLR) Optimizer for efficiently fine-tuning diffusion models, combined with a multi-scale prompting technique named Diffusive Chain-of-Thought (DCoT).
CodeExplainability and InterpretabilityComputational EfficiencyLarge Language ModelReinforcement LearningVision Language ModelMultimodalityBenchmark
🎯 What it does: Propose a LVLMs-Saliency metric based on gradient attention, and integrate Saliency-Guided Rejection Sampling with Local Coherence Reinforcement mechanisms during inference, significantly reducing hallucinations in image question answering and description generation.
Hallucination Reduction with CASAL: Contrastive Activation Steering for Amortized Learning
Wannan Yang (Meta Superintelligence Labs), Diego Garcia-Olano (Meta Superintelligence Labs)
CodeExplainability and InterpretabilityComputational EfficiencyRepresentation LearningTransformerLarge Language ModelMixture of ExpertsContrastive LearningTextMultimodality
🎯 What it does: This paper proposes a 'CASAL' training framework based on contrastive activation modulation, which directly embeds the model's knowledge boundary into weights by using a representation layer loss on a single-layer network, enabling LLMs to self-denial when encountering unknown questions and reducing hallucinations.
Hallucination-aware Intermediate Representation Edit in Large Vision-Language Models
Wei Suo (Northwestern Polytechnical University), Yanning Zhang (Northwestern Polytechnical University)
CodeExplainability and InterpretabilityRepresentation LearningVision Language ModelContrastive LearningImageText
🎯 What it does: Propose the HIRE framework, which detects and edits intermediate representations of large vision-language models to eliminate or regulate hallucinations without requiring model retraining or dual reasoning.
HardcoreLogic: Challenging Large Reasoning Models with Long-tail Logic Puzzle Games
Jingcong Liang (Fudan University), zhongyu wei
CodeLarge Language ModelTextBenchmark
🎯 What it does: Proposed a logic puzzle benchmark named HardcoreLogic, containing over 5,000 puzzles generated by three-dimensional long-tail transformations (IC, UE, UP) across multiple game genres.
Harder Is Better: Boosting Mathematical Reasoning via Difficulty-Aware GRPO and Multi-Aspect Question Reformulation
Yanqi Dai (Renmin University of China), Zhiwu Lu (Renmin University of China)
CodeOptimizationReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningTextBenchmark
🎯 What it does: Propose the MathForge framework, combining difficulty-aware Group Policy Optimization (DGPO) and multi-dimensional problem rewriting (MQR) to enhance the performance of large language models on mathematical reasoning tasks.
Harnessing Temporal Databases for Systematic Evaluation of Factual Time-Sensitive Question-Answering in LLMs
Soyeon Kim (KAIST), Steven Euijong Whang (William & Mary)
CodeExplainability and InterpretabilityTransformerLarge Language ModelTabularTime SeriesBiomedical DataBenchmarkRetrieval-Augmented GenerationChain-of-Thought
🎯 What it does: Proposed TDBench, a benchmark framework that automatically constructs time-sensitive question-answer (TSQA) pairs using a time series database, and introduced the time accuracy evaluation metric;
🎯 What it does: Proposed a conditional tabular diffusion model named HARPOON, achieving multiple constraint generation during training-free scenarios through manifold guidance.
🎯 What it does: To address the issues of imbalance and heterogeneity across datasets and within individual datasets during the fine-tuning of large language models, the Hierarchical Balancing Optimization (HBO) framework is proposed, enabling the model to autonomously adjust global and local data sampling ratios during training.
Healthcare Insurance Fraud Detection via Continual Fiedler Vector Graph Model
Yehan Zhang (South China University of Technology), Shengfeng He (Singapore Management University)
CodeAnomaly DetectionGraph Neural NetworkAuto EncoderGraphFinance Related
🎯 What it does: Designed a continuous learning graph model named ConFVG for real-time identification in medical insurance fraud detection under low-labeling and non-stationary environments, combining a Fiedler vector-guided graph autoencoder with a subgraph attention fusion module.
HEAPr: Hessian-based Efficient Atomic Expert Pruning in Output Space
Ke Li (Zhejiang University), Wenxiao Wang (Zhejiang University)
CodeComputational EfficiencyKnowledge DistillationMixture of ExpertsText
🎯 What it does: Propose a second-order information-based MoE atomic expert pruning method called HEAPr, which first splits experts into indivisible atomic experts, then estimates the importance of each atomic expert in the output space via the Fisher information matrix, achieving pruning with only two forward and one backward pass;
Hedonic Neurons: A Mechanistic Mapping of Latent Coalitions in Transformer MLPs
Tanya Chowdhury (University of Massachusetts Amherst), James Allan (University of Massachusetts Amherst)
CodeExplainability and InterpretabilityTransformerSupervised Fine-TuningText
🎯 What it does: Propose a mechanism explanation framework based on game theory, treating neurons in the Transformer MLP layer as players, leveraging synergy effects to discover and track stable neuron coalitions, revealing collaborative computational units within the model.
Helix: Evolutionary Reinforcement Learning for Open-Ended Scientific Problem Solving
Chang Su (Bosch (China) Investment Co Ltd), Jun Zhu (Tsinghua University)
CodeOptimizationTransformerLarge Language ModelReinforcement LearningPrompt EngineeringTabularBenchmark
🎯 What it does: Developed a hierarchical evolutionary reinforcement learning framework called HELIX, leveraging LLMs to achieve iterative optimization in open-ended scientific problems through experience learning, population diversity maintenance, and contextual prompting.
Helmsman: Autonomous Synthesis of Federated Learning Systems via Collaborative LLM Agents
Haoyuan Li (Eindhoven University of Technology), Aaqib Saeed (Eindhoven University of Technology)
CodeFederated LearningLarge Language ModelAgentic AIBenchmarkRetrieval-Augmented Generation
🎯 What it does: Developed a multi-agent system called Helmsman, which can automatically complete planning, modular coding, and closed-loop evaluation based on users' high-level federated learning (FL) requirements, ultimately generating a directly deployable federated learning system.
Vishal Pramanik (University of Florida), Sumit Kumar Jha (University of Florida)
CodeExplainability and InterpretabilityComputational EfficiencyTransformerLarge Language ModelText
🎯 What it does: Propose an explanation method named HETA, which provides word-level attribution for each generated word in autoregressive language models with only decoder structure, based on semantic flow, Hessian second-order sensitivity, and KL information gain.
Hey, That's My Model! Introducing Chain & Hash, An LLM Fingerprinting Technique
Mark Russinovich (Microsoft), Ahmed Salem (Microsoft)
CodeSafty and PrivacyTransformerLarge Language ModelPrompt EngineeringTextBenchmark
🎯 What it does: Propose the Chain & Hash framework, utilizing chain hashing technology to embed and detect verifiable and black-box identifiable fingerprints in LLMs.
HGNet: Scalable Foundation Model for Automated Knowledge Graph Generation from Scientific Literature
Devvrat Joshi (Imperial College London), Islem Rekik (Imperial College London)
CodeRecognitionGenerationRepresentation LearningGraph Neural NetworkTransformerLarge Language ModelTextBenchmark
🎯 What it does: Proposed a two-phase automated knowledge graph construction framework (Z-NERD for entity recognition, HGNet for hierarchical relation extraction), and released a large-scale multi-domain hierarchical relation extraction benchmark dataset SPHERE.
🎯 What it does: Proposed a training-free acceleration framework called HiCache, which enhances feature prediction accuracy by employing Hermite polynomials in the feature cache-prediction link, significantly accelerating diffusion model inference.
HiddenEcho: Mitigating Noise Amplification in Differentially Private LLMs with Hidden-State Correction
Wenhao Li (South China University of Technology), Lei Yang (South China University of Technology)
CodeFederated LearningSafty and PrivacyTransformerLarge Language ModelText
🎯 What it does: Proposed the HiddenEcho framework, achieving noise reduction and communication compression for differential privacy large language models (LLMs) through client-side noise correction and server-side hidden layer information.
HiDrop: Hierarchical Vision Token Reduction in MLLMs via Late Injection, Concave Pyramid Pruning, and Early Exit
Hao Wu (Eastern Institute of Technology), Xiaoyu Shen (Eastern Institute of Technology)
CodeComputational EfficiencyLarge Language ModelVision Language ModelMultimodality
🎯 What it does: Propose the HiDrop framework, which significantly reduces visual computational load by hierarchically pruning visual tokens in multimodal large language models.
🎯 What it does: Proposed a hierarchical entity-centric framework based on offline goal-conditioned reinforcement learning, utilizing a fact-based subgoal diffuser to generate sparse, reachable subgoals, significantly improving the success rate of long-horizon tasks.
Jiapeng Hu (North Carolina State University), Xiaorui Liu (North Carolina State University)
CodeDrug DiscoveryGraph Neural NetworkDiffusion modelBiomedical Data
🎯 What it does: Propose a hierarchical multi-scale molecular conformation generation framework named MSGEN, which first generates heavy atom skeletons as global guidance and then progressively refines hydrogen atoms or finer-grained substructures;
🎯 What it does: Studied the use of offline reinforcement learning and hierarchical value decomposition to learn whole-body robot control, achieving efficient learning on a large amount of non-expert demonstration data.
Hierarchy Decoding: A Training-free Parallel Decoding Strategy for Diffusion Large Language Models
Xiaojing Qi (Tsinghua University), Da Zheng (Ant Group)
CodeComputational EfficiencyLarge Language ModelDiffusion modelText
🎯 What it does: Propose a hierarchical recursive decoding framework, Hierarchy-dLLM, for discrete diffusion large language models (dLLMs), which significantly accelerates inference while maintaining or improving generation quality.
Hierarchy-of-Groups Policy Optimization for Long-Horizon Agentic Tasks
Shuo He (Nanyang Technological University), Bo An (Nanyang Technological University)
CodeOptimizationTransformerLarge Language ModelReinforcement LearningAgentic AIText
🎯 What it does: Propose a new hierarchical grouping + adaptive weighting advantage estimation method (HGPO) to address the advantage estimation bias caused by historical context inconsistency in long-term proxy tasks.
🎯 What it does: Transforms the visual geolocation task from image-to-image retrieval to image-to-hierarchical entity alignment using hierarchical entity embeddings in hyperbolic space;
HiFo-Prompt: Prompting with Hindsight and Foresight for LLM-based Automatic Heuristic Design
ChentongChen, Ye Fan (Northwest Polytechnical University)
CodeOptimizationTransformerLarge Language ModelPrompt Engineering
🎯 What it does: Propose HiFo‑Prompt, an automated heuristic design framework for large language models (LLMs), which achieves global control and knowledge accumulation through two modules: Foresight and Hindsight.
Yunge Li (Oakland University), Lanyu Xu (Oakland University)
CodeComputational EfficiencyTransformerImage
🎯 What it does: Proposed a local window/neighborhood attention based on the Hilbert curve, significantly improving the sparsity and computational efficiency of 2D image self-attention.
Hilbert: Recursively Building Formal Proofs with Informal Reasoning
Sumanth Varambally (University Of California San Diego), Ke Ye (University Of California San Diego)
CodeAI Code AssistantTransformerLarge Language ModelTextBenchmarkRetrieval-Augmented GenerationChain-of-Thought
🎯 What it does: Propose the HILBERT framework, integrating the informal reasoning of general-purpose LLMs with the specialized Lean prover LLM, achieving formal proofs through recursive subgoal decomposition.
Hinge Regression Tree: A Newton Method for Oblique Regression Tree Splitting
Hongyi Li (Harbin Institute of Technology), Jun Xu (Harbin Institute of Technology)
CodeOptimizationExplainability and InterpretabilityComputational EfficiencyTabular
🎯 What it does: A new Hinge Regression Tree (HRT) is proposed, achieving more efficient and interpretable piecewise decision trees by reformulating the splitting problem at each internal node as a nonlinear least squares optimization of two linear models.
🎯 What it does: This paper proposes HippoTune, which embeds a hippocampal circuit-based iterative retrieval loop in each Transformer layer. It utilizes hidden layer states as queries, performs continuous multi-round soft retrieval with feedback updates, and simulates the pattern separation and completion mechanisms of the hippocampal EC-DG-CA3-CA1 circuit, significantly reducing catastrophic forgetting in buffer-free continual learning.
HiPRAG: Hierarchical Process Rewards for Efficient Agentic Retrieval Augmented Generation
Peilin Wu (University of Texas at Dallas), Zhiyu Chen (University of Texas at Dallas)
CodeRetrievalTransformerLarge Language ModelReinforcement LearningAgentic AIPrompt EngineeringTextBenchmarkRetrieval-Augmented Generation
🎯 What it does: Investigated the efficiency of retrieval behavior in Agentic Retrieval-Augmented Generation (Agentic RAG) and proposed the HiPRAG training framework, which guides the model to improve retrieval decisions through fine-grained, hierarchical process rewards.
Histopathology-Genomics Multi-modal Structural Representation Learning for Data-Efficient Precision Oncology
Kun Wu (Beihang University), Yushan Zheng
CodeRepresentation LearningGraph Neural NetworkSpiking Neural NetworkTransformerContrastive LearningMultimodalityBiomedical Data
🎯 What it does: Propose a multi-modal structural representation learning framework, MSRL, which leverages graph structure learning (GSL) to capture associations between cases and uses real genomic information to assist inference when genomic data is missing.
HistoPrism: Unlocking Functional Pathway Analysis from Pan-Cancer Histology via Gene Expression Prediction
Susu Hu (National Center for Tumor Diseases), Stefanie Speidel (National Center for Tumor Diseases)
CodeExplainability and InterpretabilityTransformerImageBiomedical Data
🎯 What it does: Proposed the HistoPrism Transformer architecture for directly predicting cross-cancer spatial gene expression from H&E pathological images, and introduced the Gene Pathway Coherence (GPC) evaluation framework.
CodeRetrievalTransformerVision Language ModelVideoText
🎯 What it does: Propose HiTeA, a fully unsupervised and training-free long video temporal alignment framework that can localize corresponding time segments in long, uncut videos based on natural language queries.
HLD: Approximate Hierarchical Linguistic Distribution Modeling for LLM-Generated Text Detection
Rui Guo (Institute of Software, Chinese Academy of Sciences), Weiming Dong (Institute of Automation, Chinese Academy of Sciences)
CodeAnomaly DetectionExplainability and InterpretabilityComputational EfficiencyLarge Language ModelText
🎯 What it does: Proposes a framework named HLD-Detector for detecting LLM-generated text based on hierarchical language distribution, utilizing n-gram statistics and probability ratios at the word, syntactic, and semantic levels, and employing XGBoost for final judgment.
🎯 What it does: Proposed the HOTA method, which directly solves the Generalized Schrödinger Bridge using the Hamilton-Jacobi-Bellman equation and Kantarovich potential, achieving scalable trajectory optimization without explicit density modeling.
🎯 What it does: Proposes Householder-Diagonalized Linear Attention (HDLA), achieving a Diagonal-Plus-Rank-2 decay matrix through Householder matrix decomposition, and provides a general block-level parallel algorithm capable of supporting arbitrary rank decay and KV outer products; it is evaluated on tasks including large-scale pretraining, retrieval, zero-shot inference, and image classification.
Bingxiang He (Tsinghua University), Ning Ding (Tsinghua University)
CodeReinforcement Learning from Human FeedbackTransformerLarge Language ModelReinforcement LearningText
🎯 What it does: This paper systematically evaluates and theorizes the scalability of Unsupervised Verifiable Reinforcement (URLVR) in large language models (LLMs), focusing on comparing intrinsic and extrinsic reward methods. It proposes the 'Model Collapse Step' as a diagnostic metric to assess model priors and verifies the safety of intrinsic rewards in small-scale data and test-time training.
How Many Code and Test Cases Are Enough? Evaluating Test Cases Generation from a Binary-Matrix Perspective
Xianzhen Luo (Harbin Institute of Technology), Wanxiang Che (Harbin Institute of Technology)
CodeAI Code AssistantLarge Language ModelTextBenchmark
🎯 What it does: Proposed an evaluation framework based on the rank of binary matrices, designed an approximate algorithm called WrongSelect, and constructed the TC-Bench benchmark based on competition submissions.
HSG-12M: A Large-Scale Benchmark of Spatial Multigraphs from the Energy Spectra of Non-Hermitian Crystals
Xianquan Yan (National University of Singapore), Ching Hua Lee (National University of Singapore)
CodeData SynthesisRepresentation LearningGraph Neural NetworkGraphBenchmarkPhysics Related
🎯 What it does: Built the Poly2Graph automated pipeline, generating 12 million Hamiltonian spectral graphs and forming the HSG-12M dataset, and evaluated multiple GNNs on this dataset.
Hubble: a Model Suite to Advance the Study of LLM Memorization
Johnny Wei, Robin Jia (University of Southern California)
CodeSafty and PrivacyTransformerLarge Language ModelText
🎯 What it does: Propose the HUBBLE suite, which includes standard and perturbed versions of large language models, for systematically studying memory risks in LLMs.
Human Uncertainty-Aware Data Selection and Automatic Labeling in Visual Question Answering
Jian Lan (University of Munich), Thomas Seidl (University of Munich)
CodeData-Centric LearningSupervised Fine-TuningVision Language ModelMultimodality
🎯 What it does: Propose HaDola, a data selection and automatic annotation framework based on human uncertainty (HU), which can efficiently fine-tune visual question answering (VQA) models with only 5% HU-annotated data.
Human-MME: A Holistic Evaluation Benchmark for Human-Centric Multimodal Large Language Models
Yuansen Liu (National University Of Singapore), Shuicheng YAN
CodeClassificationRecognitionObject DetectionSegmentationLarge Language ModelVision Language ModelImageMultimodalityBenchmark
🎯 What it does: Constructed the Human-MME benchmark to comprehensively evaluate the fine-grained perception and high-level causal reasoning capabilities of multimodal large language models (MLLMs) on human-centric images.
🎯 What it does: Propose a human motion prediction method called HUMOF, which integrates hierarchical feature representations of human-human and human-environment interactions, along with a coarse-to-fine scale interaction reasoning module for complex dynamic interaction scenarios.
Huxley-G\"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machine
Wenyi Wang (King Abdullah University of Science and Technology), Jürgen Schmidhuber (King Abdullah University of Science and Technology)
CodeOptimizationMeta LearningAI Code AssistantLarge Language ModelText
🎯 What it does: Proposed a self-improving evaluation metric called CMP based on line families, and constructed the Huxley-Gödel Machine (HGM) guided by CMP, achieving performance comparable to humans in the automation of software engineering coding agents.
Hyper-SET: Designing Transformers via Hyperspherical Energy Minimization
Yunzhe Hu (University of Hong Kong), Dong Xu (University of Hong Kong)
CodeClassificationRestorationTransformerImage
🎯 What it does: This paper proposes a Transformer architecture HYPER-SET that shares parameters and can be recursively expanded by minimizing the energy function on a hypersphere to achieve distribution uniformity and semantic alignment.
🎯 What it does: Proposes HYPER—a foundational model for inductive link prediction on knowledge hypergraphs with arbitrary arity, supporting zero-shot reasoning for new entities and relations.
🎯 What it does: Designed and implemented SphereAR, a continuous-token image generation framework based on hyperspherical VAE and autoregressive Transformer, addressing the variance collapse problem in traditional AR models during decoding.
Ice Cream Doesn’t Cause Drowning: Benchmarking LLMs Against Statistical Pitfalls in Causal Inference
Jin Du (University of Minnesota), Jie Ding (University of Minnesota)
CodeTransformerLarge Language ModelPrompt EngineeringTabularBenchmark
🎯 What it does: Constructed the CausalPitfalls benchmark to evaluate the reliability of LLMs in statistical causal inference, covering six causal pitfalls and fifteen challenges.
🎯 What it does: Investigates the bias in information assessment caused by missing modes in multimodal learning and proposes the ICYM 2 I framework to correct training and evaluation.
IDEAL: Data Equilibrium Adaptation for Multi-Capability Language Model Alignment
Chenlin Ming (Shanghai Jiao Tong University), Conghui He (Shanghai Artificial Intelligence Laboratory)
CodeDomain AdaptationOptimizationData-Centric LearningTransformerLarge Language ModelSupervised Fine-TuningText
🎯 What it does: This paper proposes the IDEAL framework, which automatically balances data and enhances the performance of large language models (LLMs) across multiple capabilities by iteratively adjusting the proportions of multi-domain SFT training data.
Identifiability Challenges in Sparse Linear Ordinary Differential Equations
Cecilia Casolo (Technical University of Munich), Niki Kilbertus (Munich Center for Machine Learning)
CodePhysics RelatedOrdinary Differential Equation
🎯 What it does: This paper studies the identifiability problem of sparse linear ordinary differential equations (ODEs), particularly in data-driven dynamic system modeling, exploring the differences between identifiability of sparse systems and dense systems.