arXivSub Start free trial

ICLR 2026 Papers with Code β€” Page 15

International Conference on Learning Representations Β· 2207 papers

Pixel to Gaussian: Ultra-Fast Continuous Super-Resolution with 2D Gaussian Modeling

Long Peng (USTC), Zheng-Jun Zha (USTC)

CodeSuper ResolutionGaussian SplattingImageBenchmark

🎯 What it does: Propose the Pixel-to-Gaussian method, which directly reconstructs continuous high-resolution signals from low-resolution images using 2D Gaussian scattering, achieving arbitrary-scale super-resolution within a single model and significantly improving efficiency.

Pixel-Level Residual Diffusion Transformer: Scalable 3D CT Volume Generation

Zhenkai Zhang (University of Melbourne), Tom Drummond (University of Melbourne)

CodeGenerationTransformerDiffusion modelBiomedical DataComputed Tomography

🎯 What it does: This paper proposes Pixel-Level Residual Diffusion Transformer (PRDiT), which directly generates high-resolution 3D CT volumes at the voxel level through a two-stage structure (local MLP denoising + global Transformer residual learning).

Pixel-Perfect Puppetry: Precision-Guided Enhancement for Face Image and Video Editing

Yan Li (Hong Kong University of Science and Technology), Yike Guo (Hong Kong University of Science and Technology)

CodeImage TranslationGenerationConvolutional Neural NetworkDiffusion modelScore-based ModelAuto EncoderImageVideo

🎯 What it does: Proposed a unified FlowGuide framework that enables precise facial attribute editing in images and videos while preserving identity information and temporal coherence.

PixelCraft: A Multi-Agent system for High-Fidelity Visual Reasoning on Structured Images

Shuoshuo Zhang (Microsoft Research Asia), Rui Wang (Microsoft Research Asia)

CodeData SynthesisTransformerSupervised Fine-TuningAgentic AIVision Language ModelImageBenchmarkChain-of-Thought

🎯 What it does: Proposed PixelCraft, a multi-agent system that employs high-precision pixel-level tool agents and components such as a planner, critic, and image memory to perform multi-step, scalable visual reasoning on structured images (e.g., charts and geometric diagrams).

PixNerd: Pixel Neural Field Diffusion

Shuai Wang (Nanjing University), Limin Wang (Nanjing University)

CodeGenerationTransformerDiffusion modelNeural Radiance FieldImage

🎯 What it does: Directly construct the diffusion transformer PixNerd in the pixel space, achieving efficient single-stage generation through large patches and neural field decoders

Plan then Act: Bi-level CAD Command Sequence Generation

Qiangya Guo (South China University of Technology), Tianshui Chen (Zhuzhou CRRC Times Electric Co., Ltd)

CodeGenerationTransformerLarge Language ModelTextSequential

🎯 What it does: Propose the PTA two-layer CAD command sequence generation method, first using LLM to plan high-level operation plans, then generating low-level executable command sequences with a demand-aware mechanism.

Plan-Answer-Refine-on-Graph: Structured Planning and Self-Refinement for Large Language Model Reasoning on Knowledge Graphs

Yuxin Shi (Zhejiang University), Jianling Sun (Zhejiang University)

CodeTransformerLarge Language ModelTextGraphRetrieval-Augmented GenerationChain-of-Thought

🎯 What it does: Proposed the PARoG framework for large language model reasoning on knowledge graphs;

Plan-R1: Safe and Feasible Trajectory Planning as Language Modeling

Xiaolong Tang (Institute of Computing Technology, Chinese Academy of Sciences), Xilin Chen (Institute of Computing Technology, Chinese Academy of Sciences)

CodeAutonomous DrivingTransformerReinforcement LearningWorld ModelSequentialBenchmark

🎯 What it does: Propose the Plan-R1 two-stage framework: first pre-train a general trajectory predictor on expert driving data, then fine-tune with rule-based reinforcement learning to make trajectory planning explicitly adhere to safety, comfort, and rule principles.

PLANETALIGN: A Comprehensive Python Library for Benchmarking Network Alignment

Qi Yu (University of Illinois Urbana Champaign), Hanghang Tong (Texas Aamp University)

CodeGraphBenchmark

🎯 What it does: Proposed and implemented the PLANETALIGN library for unified experimentation and evaluation of network alignment methods

PlantRSR: A New Plant Dataset and Method for Reference-based Super-Resolution

Hongyang Zhou, Xu-Cheng Yin (University Of Science And Technology Beijing)

CodeSuper ResolutionKnowledge DistillationConvolutional Neural NetworkDiffusion modelGenerative Adversarial NetworkImageBenchmarkAgriculture Related

🎯 What it does: This paper proposes a reference image super-resolution method specifically for plant images and constructs a large-scale plant reference dataset called PlantRSR.

Plug-and-Play Fidelity Optimization for Diffusion Transformer Acceleration via Cumulative Error Minimization

Tong Shao (Harbin Institute of Technology), Jingyong Su (Harbin Institute of Technology)

CodeComputational EfficiencyTransformerDiffusion modelImageVideo

🎯 What it does: Propose a training-agnostic, plug-and-play cumulative error minimization plugin called CEM to optimize the caching strategy of Diffusion Transformer and improve generation quality.

Plug, Play, and Fortify: A Low-Cost Module for Robust Multimodal Image Understanding Models

Siqi Lu (National University of Defense Technology), Jianhang Yao (National University of Defense Technology)

CodeClassificationSegmentationMultimodalityBiomedical DataMagnetic Resonance Imaging

🎯 What it does: Propose a frequency-domain based modality preference metric (FRM), and design the MWAM module using FRM to dynamically balance gradients/losses of the multimodal network during training, thereby enhancing robustness in scenarios with missing modalities.

PM-KVQ: Progressive Mixed-precision KV Cache Quantization for Long-CoT LLMs

Tengxuan Liu, Yu Wang (Infinigence-Ai)

CodeComputational EfficiencyTransformerLarge Language ModelTextChain-of-Thought

🎯 What it does: This paper proposes a post-training KV cache quantization method called Progressive Mixed-Precision KV Cache Quantization (PM-KVQ), specifically designed for large language models engaged in long-chain reasoning (long-CoT).

PMDformer: Patch-Mean Decoupling Information Transformer for Long-term Forecasting

Ao Hu (Southwestern University of Finance and Economics), Zenglin Xu (Shanghai Academy of AI for Science)

CodeTransformerTime SeriesBenchmark

🎯 What it does: Propose the PMDformer model, addressing the challenge of shape similarity recognition caused by scale differences in long-term time series prediction through modules such as patch-mean decoupling, proximal variable attention, and trend restoration attention, thereby achieving more accurate multivariate forecasting.

PoinnCARE: Hyperbolic Multi-Modal Learning for Enzyme Classification

Kun XIE, Biaobin Jiang (Tencent)

CodeClassificationGraph Neural NetworkTransformerLarge Language ModelMultimodalityBiomedical DataBenchmark

🎯 What it does: Proposes PoinnCARE, a hyperbolic (Poincaré ball) embedding framework based on bimodal data (sequence, structure, active site), utilizing graph diffusion and cross-modal alignment for enzyme EC number prediction.

Point-Focused Attention Meets Context-Scan State Space: Robust Biological Visual Perception for Point Cloud Representation

Kanglin Qu (Nanjing University of Aeronautics and Astronautics), Yuanhao Sun (Beijing University of Posts and Telecommunications)

CodeClassificationSegmentationTransformerPoint Cloud

🎯 What it does: Proposes PointLearner, a point cloud representation learning network that mimics the foveal vision of the human retina, capable of simultaneously capturing local details and global context.

Point2RBox-v3: Self-Bootstrapping from Point Annotations via Integrated Pseudo-Label Refinement and Utilization

Teng Zhang (Shanghai Jiao Tong University), Xue Yang (East China Normal University)

CodeObject DetectionConvolutional Neural NetworkImagePoint Cloud

🎯 What it does: Proposes a self-supervised rotation object detection framework named Point2RBox-v3 based on point annotations, integrating dynamic pseudo-labels and mask augmentation.

PoLi-RL: A Point-to-List Reinforcement Learning Framework for Conditional Semantic Textual Similarity

Zixin Song (Tsinghua University), Chunping Li (Tsinghua University)

CodeRetrievalTransformerLarge Language ModelReinforcement LearningText

🎯 What it does: This study proposes PoLi-RLβ€”a two-phase reinforcement learning (RL) framework for conditional semantic text similarity (C-STS)β€”which gradually transitions from point-wise rewards to a hybrid strategy combining point-wise, pair-wise, and list-wise rewards during training, leveraging LLM cross-encoders.

Policy Likelihood-based Query Sampling and Critic-Exploited Reset for Efficient Preference-based Reinforcement Learning

Jongkook Heo (Korea University), Seoung Bum Kim (Korea University)

CodeReinforcement Learning from Human FeedbackReinforcement Learning

🎯 What it does: Proposes PoLiCER, combining policy-likelihood-based query sampling and critic-based reset to enhance the efficiency and performance of preference-based reinforcement learning.

Polynomial, trigonometric, and tropical activations

Ismail Khalfaoui-Hassani (Forschungszentrum JΓΌlich), Stefan Kesselheim (Forschungszentrum JΓΌlich)

CodeOptimizationComputational EfficiencyRepresentation LearningConvolutional Neural NetworkTransformerImageText

🎯 What it does: Propose learnable activation functions based on orthogonal function bases (Hermite, Fourier, tropical polynomials) and corresponding variance-preserving initialization, and verify their trainability on large-scale vision and language tasks.

PolySHAP: Extending KernelSHAP with Interaction-Informed Polynomial Regression

Fabian Fumagalli (Bielefeld University), Christopher Musco (New York University)

CodeExplainability and InterpretabilityImageTextTabular

🎯 What it does: Propose the PolySHAP method, using polynomial regression (including higher-order interaction terms) to approximate the game function, thereby more accurately estimating Shapley values;

PolySkill: Learning Generalizable Skills Through Polymorphic Abstraction For Continual Learning

Simon Yu (Northeastern University), Peng Qi (Uniphore)

CodeMeta LearningTransformerLarge Language ModelAgentic AIText

🎯 What it does: Propose the PolySkill framework, enabling Web agents to learn generalizable and composable skills through polymorphic abstraction;

PonderLM: Pretraining Language Models to Ponder in Continuous Space

Boyi Zeng (Shanghai Jiao Tong University), Zhouhan Lin (Shanghai Jiao Tong University)

CodeRepresentation LearningTransformerLarge Language ModelTextBenchmark

🎯 What it does: In each token generation step of language models, a self-β€˜pondering’ (deep thinking) mechanism is formed by obtaining continuous embeddings through multiple forward passes and weighting with the prediction distribution.

PoseX: AI Defeats Physics-based Methods on Protein Ligand Cross-Docking

Yize Jiang (Microcyto), Junhong Liu (Microcyto)

CodeDrug DiscoveryProtein Structure PredictionDiffusion modelBiomedical DataBenchmark

🎯 What it does: Constructed the PoseX benchmark, collected 718 self-docking and 1312 cross-docking datasets, conducted unified evaluations of 23 physics-based, AI docking, and AI co-folding methods, and proposed an energy minimization relaxation module.

PoSh: Using Scene Graphs to Guide LLMs-as-a-Judge for Detailed Image Descriptions

Amith Ananthram (Columbia University), Kathleen McKeown (Ucla)

CodeExplainability and InterpretabilityTransformerLarge Language ModelImageTextGraphBenchmark

🎯 What it does: Propose a fine-grained image description evaluation metric called POSH based on scene graphs, and construct a fine-grained evaluation benchmark named DOCENT containing artworks.

PPE: Positional Preservation Embedding for Token Compression in Multimodal Large Language Models

Mouxiao Huang (Huawei Technologies), Xinghao Chen (Huawei Technologies)

CodeCompressionTransformerLarge Language ModelVision Language ModelMultimodality

🎯 What it does: Propose Positional Preservation Embedding (PPE) to preserve spatial and temporal positional information when compressing visual tokens in multimodal large language models.

Practical estimation of the optimal classification error with soft labels and calibration

Ryota Ushio (University of Tokyo), Masashi Sugiyama (RIKEN AIP)

CodeClassificationImageText

🎯 What it does: This paper proposes a method to estimate the optimal classification error rate under a binary classification setup, extending previous work that utilized soft labels to estimate Bayesian error.

Pre-training Limited Memory Language Models with Internal and External Knowledge

Linxi Zhao (Cornell University), Jennifer J. Sun (Cornell University)

CodeComputational EfficiencyRepresentation LearningTransformerLarge Language ModelTextGraphRetrieval-Augmented Generation

🎯 What it does: Propose a new pre-trained language model, LMLM, which stores entity-level facts in an external database and during pre-training inserts lookup calls with loss masking over retrieved values, enabling the model to learn how to query rather than memorize facts.

Precise and Interpretable Editing of Code Knowledge in Large Language Models

Min Xue (Heidelberg University), Artur Andrzejak (Heidelberg University)

CodeExplainability and InterpretabilityAI Code AssistantTransformerLarge Language ModelTextBenchmark

🎯 What it does: Propose TransCoder-based Precise Editing (TCPE) for code knowledge editing and design a functional equivalence evaluation benchmark KECode.

Predicting Kernel Regression Learning Curves from Only Raw Data Statistics

Dhruva Karkada (UC Berkeley), James B Simon

CodeRepresentation LearningImage

🎯 What it does: This paper proposes and verifies the Hermite Eigenstructure Ansatz (HEA), which can predict the learning curves of rotation-invariant kernel ridge regression (KRR) using only the data covariance matrix and the Hermite decomposition of the objective function.

Prediction with Expert Advice under Local Differential Privacy

Ben Jacobsen (University of Wisconsin Madison), Kassem Fawaz (University of Wisconsin Madison)

CodeSafty and PrivacyTabularBiomedical DataElectronic Health Records

🎯 What it does: This study improves the expert advice prediction problem under local differential privacy (LDP) constraints, proposing two algorithms: RW-AdaBatch and RW-Meta;

Predictive Differential Training Guided by Training Dynamics

Fanqi Wang (University of Tennessee, Knoxville), Igor Mezic (University of California, Santa Barbara)

CodeOptimizationComputational EfficiencyConvolutional Neural NetworkTransformerImage

🎯 What it does: Proposed and implemented a pluggable Predictive Differential Training (PDT) framework that selectively accelerates weight updates during training using a Koopman prediction model.

PrefDisco: Benchmarking Proactive Personalized Reasoning

Shuyue Stella Li (University of Washington), Yulia Tsvetkov (University of Washington)

CodeLarge Language ModelTextBenchmark

🎯 What it does: Designed and implemented the PREFDISCO evaluation framework, transforming static reasoning benchmarks into interactive personalized assessments, and defined the PREFALIGN metric to measure the alignment between responses and user preferences.

Preference Leakage: A Contamination Problem in LLM-as-a-judge

Dawei Li (Arizona State University), huan liu

CodeExplainability and InterpretabilityData-Centric LearningTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningPrompt EngineeringTextBenchmark

🎯 What it does: This paper reveals the preference leakage problem in LLM-as-a-judge systems, where the association between the data-generating LLM and the evaluation LLM leads to biased judgments favoring the student model.

Preference-based Policy Optimization from Sparse-reward Offline Dataset

Wenjie Qiu (Rutgers University), He Zhu (Rutgers University)

CodeOptimizationReinforcement LearningContrastive LearningBenchmark

🎯 What it does: Developed the PREFORL algorithm, using contrastive preference learning to optimize policies on sparse reward offline data.

Premise Selection for a Lean Hammer

Thomas Zhu (Carnegie Mellon University), Sean Welleck (Carnegie Mellon University)

CodeAI Code AssistantTransformerContrastive LearningText

🎯 What it does: Proposed LEARNPREMISE, a premise selector based on neural networks, and combined it with tools such as Lean-auto, Aesop, and Duper to build the first end-to-end Lean hammer called LEANHAMMER;

Presenting a Paper is an Art: Self-Improvement Aesthetic Agents for Academic Presentations

Chengzhi Liu (University of California, Santa Barbara), Xin Eric Wang (University of California, Santa Barbara)

CodeGenerationReinforcement LearningAgentic AIVision Language ModelImageVideoTextMultimodalityBenchmark

🎯 What it does: Proposes the EvoPresent framework, which automates the generation of academic presentations and enhances content and visual quality through self-improvement.

Preserve and Sculpt: Manifold-Aligned Fine-tuning of Vision-Language Models for Few-Shot Learning

Dexia Chen (Sun Yat-sen University), Ruixuan Wang (Sun Yat-sen University)

CodeClassificationRepresentation LearningMeta LearningTransformerSupervised Fine-TuningVision Language ModelContrastive LearningMultimodality

🎯 What it does: Propose a fine-tuning framework called MPS-Tuning based on semantic manifold alignment and sculpting, which can improve the performance of few-shot vision-language tasks while preserving the knowledge of pre-trained models.

Preserving Forgery Artifacts: AI-Generated Video Detection at Native Scale

Zhengcen Li (Harbin Institute of Technology), Jingyong Su (Harbin Institute of Technology)

CodeData SynthesisAnomaly DetectionTransformerSupervised Fine-TuningVision Language ModelVideo

🎯 What it does: Studied AI-generated video detection, proposing native spatiotemporal scale processing for videos and constructing a large-scale multi-generator dataset.

Prima.cpp: Fast 30-70B LLM Inference on Heterogeneous and Low-Resource Home Clusters

Zonghang Li (MBZUAI), Xue Liu (MBZUAI)

CodeComputational EfficiencyTransformerLarge Language ModelText

🎯 What it does: This paper proposes prima.cpp, a distributed LLM inference system designed for home clusters, capable of running Llama and other large models with 30–70B parameters under extreme conditions such as mixed CPU/GPU, insufficient memory/VRAM, slow disks, and Wi-Fi.

Principled Fast and Meta Knowledge Learners for Continual Reinforcement Learning

Ke Sun (University of Alberta), Linglong Kong (University of Alberta)

CodeMeta LearningReinforcement LearningBenchmark

🎯 What it does: Propose a dual-learner framework (fast learner and meta-learner), where the fast learner enables knowledge transfer and the meta-learner enables knowledge integration to address knowledge transfer and catastrophic forgetting in continuous reinforcement learning.

Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective

Jingyang Ou (Renmin University of China), Chongxuan Li (Renmin University of China)

CodeGenerationTransformerLarge Language ModelReinforcement LearningDiffusion modelText

🎯 What it does: A full-sequence level reinforcement learning framework (ESPO) is proposed for diffusion-based large language models (dLLMs), which constructs a sequence-level advantage function and KL regularizer to perform post-training optimization by using ELBO as a computable proxy for sequence likelihood.

Prior-aware and Context-guided Group Sampling for Active Probabilistic Subsampling

Beomgu Kang (Korea University), Hyunseok Seo (Korea University)

CodeOptimizationComputational EfficiencyRepresentation LearningConvolutional Neural NetworkRecurrent Neural NetworkImageBiomedical DataMagnetic Resonance Imaging

🎯 What it does: Proposed a grouped active sampling method named PGA-DPS that combines prior information and context guidance to achieve efficient subsampling across different tasks.

Prior-based Noisy Text Data Filtering: Fast and Strong Alternative For Perplexity

Yeongbin Seo (Yonsei University), Jinyoung Yeo (Yonsei University)

CodeComputational EfficiencyData-Centric LearningText

🎯 What it does: This paper proposes a text data filtering method based on lexical priors, using word frequency statistics instead of traditional PPL filtering;

Prior-free Tabular Test-time Adaptation

Rundong He (Hong Kong Polytechnic University), Jieming Shi (Hong Kong Polytechnic University)

CodeDomain AdaptationTabularBenchmark

🎯 What it does: Propose a table-based test-time adaptation method PFT3A without source prior and source data

PriorGuide: Test-Time Prior Adaptation for Simulation-Based Inference

Yang Yang (University of Helsinki), Luigi Acerbi (Aalto University)

CodeDomain AdaptationDiffusion modelScore-based ModelTabularBenchmarkStochastic Differential Equation

🎯 What it does: Proposed a framework called PriorGuide, which enables diffusion posterior inference to adapt to new prior distributions during testing without retraining, based on a pre-trained diffusion model.

PRISON: Unmasking the Criminal Potential of Large Language Models

Xinyi Wu (Fudan University), Min Yang (Fudan University)

CodeSafty and PrivacyTransformerLarge Language ModelTextChain-of-Thought

🎯 What it does: Proposed the PRISON framework, which evaluates the criminal potential and detection capability of large language models (LLMs) in real criminal scenarios from three perspectives.

Private Rate-Constrained Optimization with Applications to Fair Learning

Mohammad Yaghini (University of Toronto), Nicolas Papernot (University of Toronto)

CodeOptimizationSafty and PrivacyConvolutional Neural NetworkImageTabular

🎯 What it does: Developed an algorithm called RaCO-DP that can optimize machine learning models with rate constraints while satisfying differential privacy (DP);

Probabilistic Kernel Function for Fast Angle Testing

Kejing Lu (Yamanashi University), Yoshiharu Ishikawa (Nagoya University)

CodeRetrievalComputational EfficiencyImageText

🎯 What it does: Propose two probability kernel functions K1S and K2S based on reference angles for angle comparison and angle threshold determination in high-dimensional Euclidean space, and design KS1 projection method and KS2 routing test based on this.

Probing in the Dark: State Entropy Maximization for POMDPs

Yonatan Ashlag (Technion Israel Institute of Technology), Kfir Yehuda Levy

CodeReinforcement LearningBenchmark

🎯 What it does: Proposes a method for reward-free pre-training in partially observable Markov decision processes (POMDP) by maximizing the entropy of predicted latent states, and implements the LatEnt algorithm and PROBE benchmark;

Probing to Refine: Reinforcement Distillation of LLM Reasoners via Explanatory Inversion

Zhen Tan (Arizona State University), huan liu

CodeExplainability and InterpretabilityKnowledge DistillationTransformerLarge Language ModelSupervised Fine-TuningReinforcement LearningText

🎯 What it does: Propose the ExGRPO framework, which combines Explanatory Inversion (EI) to generate diverse explanatory probing questions, and incorporates dialogue structure utility rewards in reinforcement learning to achieve knowledge distillation and reasoning capability enhancement for small-scale LLMs.

Process-Level Trajectory Evaluation for Environment Configuration in Software Engineering Agents

Jiayi Kuang (Sun Yat-sen University), Philip S. Yu (University Of Illinois Chicago)

CodeData SynthesisAI Code AssistantTransformerLarge Language ModelAgentic AITextBenchmark

🎯 What it does: Propose EnConda-Bench, a process-oriented environment configuration evaluation framework, and implement automated data generation with Docker validation

ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

Zhilin Wang (NVIDIA), Yi Dong (NVIDIA)

CodeTransformerLarge Language ModelPrompt EngineeringTextBenchmarkFinance RelatedPhysics RelatedChain-of-Thought

🎯 What it does: Create ProfBench benchmark, covering 7,347 professional domain tasks with corresponding evaluation criteria, and assess LLMs' capabilities in generation and judgment.

Projected Coupled Diffusion for Test-Time Constrained Joint Generation

Hao Luan (National University of Singapore), Chun Kai Ling (National University of Singapore)

CodeGenerationRobotic IntelligenceDiffusion modelScore-based ModelImageStochastic Differential Equation

🎯 What it does: Proposed Projected Coupled Diffusion (PCD), a framework that generates samples satisfying hard constraints by jointly utilizing multiple pre-trained diffusion models during testing

Prompt and Parameter Co-Optimization for Large Language Models

Xiaohe Bo (Renmin University of China), Zhenhua Dong (Renmin University of China)

CodeOptimizationTransformerLarge Language ModelSupervised Fine-TuningPrompt EngineeringText

🎯 What it does: Propose the MetaTuner framework to achieve jointly collaborative optimization of prompt tuning and model fine-tuning, thereby enhancing the task performance of large language models (LLMs).

Prompt-MII: Meta-Learning Instruction Induction for LLMs

Emily Xiao (Carnegie Mellon University), Graham Neubig (Carnegie Mellon University)

CodeMeta LearningTransformerReinforcement LearningPrompt EngineeringText

🎯 What it does: Train a meta-learning instruction generator to enable LLMs to generate concise task instructions in a single forward pass, replacing traditional in-context learning or manual prompts.

PromptHub: Enhancing Multi-Prompt Visual In-Context Learning with Locality-Aware Fusion, Concentration and Alignment

Tianci Luo (Tsinghua University), Shu-Tao Xia (Tsinghua University)

CodeObject DetectionSegmentationTransformerPrompt EngineeringAuto EncoderContrastive LearningImageRetrieval-Augmented Generation

🎯 What it does: Propose the PromptHub framework in visual context learning, using multi-prompt fusion to enhance performance in image segmentation, detection, and coloring tasks.

ProofFlow: A Dependency Graph Approach to Faithful Proof Autoformalization

Rafael Medeiros Cabral (Huawei Celia Team), SHEN XIN (Huawei Celia Team)

CodeAI Code AssistantTransformerLarge Language ModelTextGraphBenchmark

🎯 What it does: Developed a three-stage pipeline, PROOFFLOW, to convert natural language proofs into structured Lean code, and introduced the PROOFSCORE evaluation metric along with the PROOFFLOWBENCH dataset.

Propaganda AI: An Analysis of Semantic Divergence in Large Language Models

Nay Myat Min (Singapore Management University), Jun Sun (Singapore Management University)

CodeExplainability and InterpretabilityAdversarial AttackTransformerLarge Language ModelSupervised Fine-TuningPrompt EngineeringText

🎯 What it does: Studied a black-box auditing framework called RAVEN for detecting semantic biases and propaganda consistency in large language models induced by high-level concepts (e.g., public figures, ideologies).

PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approach

Udari Madhushani Sehwag (Scale AI), Furong Huang (University of Maryland)

CodeSafty and PrivacyLarge Language ModelAgentic AITextBenchmark

🎯 What it does: Proposed PropensityBench, an agentic method to evaluate the propensity of large language models when using high-risk tools, covering 5,874 tasks across four high-risk domains: cybersecurity, self-replication, life safety, and chemical safety.

Proper Velocity Neural Networks

Ziheng Chen (University of Trento), Nicu Sebe (University of Trento)

CodeClassificationImageGraphBiomedical Data

🎯 What it does: Proposes a neural network (PVNN) built using an unconstrained Proper Velocity (PV) space, constructing PV versions of polynomial logistic regression (MLR), fully connected layers (FC), convolutional layers, activation functions, and normalization layers by deriving its complete Riemannian geometry toolbox;

ProPerSim: Developing Proactive and Personalized AI Assistants through User-Assistant Simulation

Jiho Kim (Korea Advanced Institute of Science and Technology), Edward Choi (Seoul National University)

CodeRecommendation SystemData-Centric LearningReinforcement Learning from Human FeedbackTransformerLarge Language ModelSupervised Fine-TuningTextBenchmarkRetrieval-Augmented Generation

🎯 What it does: Investigated an AI assistant that simultaneously exhibits proactivity and personalization, learning from user agents in simulated home environments to provide personalized recommendations.

ProSafePrune: Projected Safety Pruning for Mitigating Over-Refusal in LLMs

Zijun Chen (Hefei University of Technology), Richang Hong (Hefei University of Technology)

CodeSafty and PrivacyComputational EfficiencyTransformerLarge Language ModelTextBenchmark

🎯 What it does: To address the problem of over-rejection in large language models (LLMs), the ProSafePrune low-rank parameter pruning method is proposed, which directly removes excessive 'harmful encoding' within the model for pseudo-harmful instructions, thereby reducing false rejection rates while maintaining safe rejection capabilities.

ProstaTD: Bridging Surgical Triplet from Classification to Fully Supervised Detection

Yiliang Chen, Jing Qin (Hong Kong Polytechnic University)

CodeObject DetectionKnowledge DistillationConvolutional Neural NetworkTransformerVideoBiomedical DataBenchmark

🎯 What it does: This study proposes the ProstaTD dataset and the TDnet benchmark model, constructing the first fully supervised triplet detection resource in surgery, covering complete videos of multi-institutional robotic prostatectomy, precise bounding boxes, and temporary boundaries;

ProTDyn: A Foundation Protein Language Model for Thermodynamics and Dynamics Generation

Yikai Liu (Purdue University), Guang Lin (Purdue University)

CodeGenerationDrug DiscoveryTransformerLarge Language ModelTime SeriesBiomedical Data

🎯 What it does: What was done: Proposed ProTDyn, a unified multi-task autoregressive Transformer model capable of generating protein thermodynamic equilibrium conformation ensembles and multi-scale dynamic trajectories within the same framework.

Protein Structure Tokenization via Geometric Byte Pair Encoding

Michael Sun (MIT), Marinka Zitnik (Apple)

CodeRepresentation LearningProtein Structure PredictionTransformerBiomedical Data

🎯 What it does: Designed and implemented a geometry-based byte pair encoding (GEOBPE) that segments continuous protein backbones into discrete hierarchical structural tokens.

ProteinAE: Protein Diffusion Autoencoders for Structure Encoding

Shaoning Li (CUHK), Pheng-Ann Heng (CUHK)

CodeProtein Structure PredictionTransformerDiffusion modelFlow-based ModelAuto EncoderBiomedical Data

🎯 What it does: Proposed the PROTEINAE autoencoder, which directly maps protein backbone coordinates to a continuous compact latent space, and trained a latent diffusion model for structure generation based on this.

ProtoKV: Long-context Knowledges Are Already Well-Organized Before Your Query

Zhiyuan Yu (Nanjing University), Sanglu Lu (Nanjing University)

CodeCompressionTransformerText

🎯 What it does: Studied the token distribution during the prefilling stage of large language models when compressing Key-Value (KV) caches, discovering that a small number of special semantic anchor points (SAT) cluster in the key embedding space, and proposed the ProtoKV method for KV cache compression based on this observation.

ProtoTS: Learning Hierarchical Prototypes for Explainable Time Series Forecasting

Ziheng Peng (Renmin University of China), Liang Sun (Alibaba Group)

CodeExplainability and InterpretabilityTime SeriesFinance Related

🎯 What it does: Proposes the ProtoTS framework, which interprets time series prediction using hierarchical prototypes and extracts interactive information from heterogeneous inputs through multi-channel embeddings and bottleneck fusion;

Provable and Practical In-Context Policy Optimization for Self-Improvement

Tianrun Yu (Brigham Young University), Weitong Zhang (University of North Carolina at Chapel Hill)

CodeOptimizationTransformerLarge Language ModelReinforcement LearningTextBenchmarkChain-of-Thought

🎯 What it does: Proposes the In-Context Policy Optimization (ICPO) framework and its practical algorithm ME-ICPO, which achieve test-time scaling and answer improvement for large language models (LLMs) through self-reflection and self-evaluation rewards during inference.

Provably Accelerated Imaging with Restarted Inertia and Score-based Image Priors

Marien Renaud (University of Bordeaux), Yu Sun (Johns Hopkins University)

CodeRestorationSuper ResolutionOptimizationScore-based ModelImageMagnetic Resonance ImagingOrdinary Differential Equation

🎯 What it does: Proposed the RISP (Restarted Inertia with Score-based Priors) algorithm, which accelerates the iterative convergence of RED while maintaining score-based image priors through restarted inertia.

Proximal Diffusion Neural Sampler

Wei Guo (Georgia Institute of Technology), Yongxin Chen (Georgia Institute of Technology)

CodeGenerationOptimizationDiffusion modelScore-based ModelMultimodalityGraphBiomedical DataPhysics RelatedStochastic Differential Equation

🎯 What it does: Propose the Proximal Diffusion Neural Sampler (PDNS), which achieves step-by-step learning of diffusion neural samplers by applying proximal point methods in the path measure space, and provides a concrete implementation based on weighted denoising cross-entropy (WDCE).

ProxyAttn: Guided Sparse Attention via Representative Heads

Yixuan Wang (Harbin Institute of Technology), Wanxiang Che (Harbin Institute of Technology)

CodeComputational EfficiencyTransformerTextBenchmark

🎯 What it does: Propose a training-agnostic sparse attention method called ProxyAttn, which uses representative heads to compress head dimensions and estimate attention scores for all heads, achieving more precise block importance assessment.

ProxyThinker: Test-Time Guidance through Small Visual Reasoners

Zilin Xiao (Rice University), Vicente Ordonez (Rice University)

CodeComputational EfficiencyKnowledge DistillationReinforcement LearningImageMultimodalityChain-of-Thought

🎯 What it does: Proposed a reasoning-guidance method called PROXYTHINKER, which adjusts the output distribution of large models by utilizing the contrastive differences from RFT reward reinforcement of small models, thereby enhancing visual reasoning capabilities without additional training.

Prune Redundancy, Preserve Essence: Vision Token Compression in VLMs via Synergistic Importance-Diversity

Zhengyao Fang (Harbin Institute of Technology), Wenjie Pei (Harbin Institute of Technology)

CodeCompressionVision Language ModelMultimodality

🎯 What it does: Proposes a training-free, task-agnostic visual token compression framework called PRUNESID, which compresses visual tokens in Vision-Language Models (VLMs) through a two-stage process.

Pruning as a Cooperative Game: Surrogate-Assisted Layer Contribution Estimation for Large Language Models

Xuan Ding (Chinese University of Hong Kong), Yao Zhu (Zhejiang University)

CodeExplainability and InterpretabilityComputational EfficiencyTransformerLarge Language ModelText

🎯 What it does: Proposes a hierarchical pruning framework based on cooperative game theory, which dynamically identifies and prunes layers with the least contribution to model performance by approximating layer Shapley values using a lightweight proxy network and hierarchical Monte Carlo mask sampling.

Pruning Long Chain-of-Thought of Large Reasoning Models via Small-Scale Preference Optimization

Bin Hong (University of Science and Technology of China), Zhenya Huang (University of Science and Technology of China)

CodeOptimizationComputational EfficiencyReinforcement Learning from Human FeedbackTransformerLarge Language ModelSupervised Fine-TuningTextChain-of-Thought

🎯 What it does: Reduce the output length of large-scale inference models while maintaining inference performance through small-scale preference optimization.

PSDNorm: Temporal Normalization for Deep Learning in Sleep Staging

Theo Gnassounou, Alexandre Gramfort (Γ‰cole Polytechnique)

CodeClassificationConvolutional Neural NetworkTransformerTime SeriesBiomedical Data

🎯 What it does: Proposes the PSDNorm layer, achieving more robust normalization in sleep staging tasks by aligning the power spectral density of feature maps through Monge mapping in deep learning models.

PT$^2$-LLM: Post-Training Ternarization for Large Language Models

Xianglong Yan (Shanghai Jiao Tong University), Yulun Zhang (Shanghai Jiao Tong University)

CodeCompressionComputational EfficiencyLarge Language ModelText

🎯 What it does: Propose PT-LLM, a ternary framework that compresses large language models to approximately 1.58-bit without retraining, significantly reducing memory and computational costs while maintaining high accuracy.

PTQ4ARVG: Post-Training Quantization for AutoRegressive Visual Generation Models

Xuewen Liu (Institute of Automation, Chinese Academy of Sciences), Qingyi Gu (Institute of Automation, Chinese Academy of Sciences)

CodeGenerationComputational EfficiencyTransformerImage

🎯 What it does: Studied post-training quantization methods for autoregressive visual generation models (ARVG), and proposed the PTQ4ARVG framework.

PU-BENCH: A UNIFIED BENCHMARK FOR RIGOROUS AND REPRODUCIBLE PU LEARNING

Qiuyi Chen (Xi'an Jiaotong-Liverpool University), Wei Wang (Xi'an Jiaotong-Liverpool University)

CodeClassificationImageTextMultimodalityTabularBenchmark

🎯 What it does: Proposes PU-Bench, a unified and reproducible benchmark for learning with incomplete positive and negative labels, integrating 18 advanced algorithms, 8 multimodal datasets, and providing standardized data generation, training, and evaluation processes.

Purrception: Variational Flow Matching for Vector-Quantized Image Generation

Răzvan-Andrei Matișan (UvA-Bosch Delta Lab University of Amsterdam), Floor Eijkelboom (UvA-Bosch Delta Lab University of Amsterdam)

CodeGenerationRepresentation LearningTransformerDiffusion modelFlow-based ModelAuto EncoderImage

🎯 What it does: Proposes a variable flow matching method named Purrception for generating high-resolution images in the vector quantized (VQ) latent space.

Pursuing Minimal Sufficiency in Spatial Reasoning

Yejie Guo (Shanghai Jiao Tong University), Ming-Hsuan Yang

CodeRepresentation LearningTransformerLarge Language ModelAgentic AIVision Language ModelMultimodalityBenchmarkChain-of-Thought

🎯 What it does: Proposed and implemented a zero-training dual-agent framework called MSSR, which first extracts 3D scene information via the Perception Agent, then constructs a minimal sufficient set (MSS) through iterative refinement and request mechanisms by the Reasoning Agent to answer spatial reasoning questions.

Pusa V1.0: Unlocking Temporal Control in Pretrained Video Diffusion Models via Vectorized Timestep Adaptation

Yaofang Liu (City University of Hong Kong), Jean-michel Morel

CodeGenerationTransformerSupervised Fine-TuningDiffusion modelVideoTextBenchmarkOrdinary Differential Equation

🎯 What it does: Proposed Pusa V1.0, which achieves fine-grained temporal control on pre-trained text-to-video (T2V) diffusion models through vectorized time step adaptation (VTA), supporting multi-tasks such as zero-shot image-to-video generation, start-end frame control, and video extension.

PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehunts

Hengzhi Li (MIT), Paul Pu Liang (MIT)

CodeTransformerLarge Language ModelSupervised Fine-TuningVision Language ModelMultimodalityBenchmarkChain-of-Thought

🎯 What it does: Proposes the PUZZLEWORLD benchmark, containing 667 real-world puzzlehunt problems, providing fine-grained multi-modal inputs, cognitive skill labels, and step-by-step reasoning trajectories to evaluate open-ended multi-modal reasoning capabilities.

PYRREGULAR: A Unified Framework for Irregular Time Series, with Classification Benchmarks

Francesco Spinnato (University of Pisa), Cristiano Landi (University of Pisa)

CodeClassificationTime SeriesBenchmark

🎯 What it does: Proposed the pyrregular framework, which unifies the storage and processing of irregular time series using a single sparse array format, and constructed a standard repository of 34 naturally irregular time series classification datasets, followed by a systematic benchmark evaluation of 12 mainstream classifiers.

Q-RAG: Long Context Multi‑Step Retrieval via Value‑Based Embedder Training

Artyom Sorokin (AXXX), Evgeny Burnaev (AXXX)

CodeRetrievalReinforcement LearningTextBenchmarkRetrieval-Augmented Generation

🎯 What it does: Proposed and trained Q-RAG, an efficient model that performs multi-step retrieval in the embedding space using reinforcement learning.

QeRL: Beyond Efficiency - Quantization-enhanced Reinforcement Learning for LLMs

Wei Huang (NVIDIA), Yukang Chen (NVIDIA)

CodeComputational EfficiencyLarge Language ModelReinforcement LearningText

🎯 What it does: Propose a framework named QeRL, combining 4-bit NVFP4 quantization with LoRA fine-tuning to achieve efficient training of large language models in reinforcement learning tasks.

QLCoder: A Query Synthesizer For Static Analysis of Security Vulnerabilities

Claire Wang (University of Pennsylvania), Mayur Naik (University of Pennsylvania)

CodeAI Code AssistantTransformerLarge Language ModelAgentic AITextRetrieval-Augmented Generation

🎯 What it does: Propose a LLM-based agent framework called QLCoder, which can automatically generate and verify CodeQL static analysis queries from CVE descriptions.

QLIP: A Dynamic Quadtree Vision Prior Enhances MLLM Performance Without Retraining

Kyle R. Chickering (University of California Davis), Muhao Chen (University of California Davis)

CodeVision Language ModelContrastive LearningImageMultimodality

🎯 What it does: Propose QLIP, a lightweight, pluggable quadtree-based improvement scheme for the CLIP vision encoder, enabling MLLM to perform reasoning on images of arbitrary resolutions without requiring retraining.

Quadratic Direct Forecast for Training Multi-Step Time-Series Forecast Models

Eric Wang, Zhouchen Lin (Peking University)

CodeOptimizationMeta LearningTransformerTime Series

🎯 What it does: Proposed a quadratic form weighted learning objective along with the MetaDF/QDF algorithms, using an adaptive weight matrix to enhance the training of multi-step time series prediction models

Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead

Feiyang Kang (FAIR at Meta), Newsha Ardalani (FAIR at Meta)

CodeTransformerSupervised Fine-TuningReinforcement LearningText

🎯 What it does: Investigate the predictability of RL after SFT, evaluate whether SFT scores can reliably predict RL outcomes, and propose new evaluation metrics;

Quant-dLLM: Post-Training Extreme Low-Bit Quantization for Diffusion Large Language Models

Tianao Zhang (Shanghai Jiao Tong University), Yulun Zhang (Shanghai Jiao Tong University)

CodeGenerationComputational EfficiencyTransformerLarge Language ModelDiffusion modelText

🎯 What it does: Provide a 2-bit weight quantization scheme for diffusion large language models

Quantifying Cross-Attention Interaction in Transformers for Interpreting TCR-pMHC Binding

Jiarui Li (Tulane University), Ramgopal R. Mettu (Tulane University)

CodeExplainability and InterpretabilityDrug DiscoveryTransformerBiomedical DataBenchmark

🎯 What it does: Propose a post-hoc explanation method called QCAI to quantify cross-attention in Transformer decoders, aiding in the interpretation of TCR-pMHC binding prediction models.

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning

Junkang Wu (University of Science and Technology of China), Xiangnan He (University of Science and Technology of China)

CodeLarge Language ModelReinforcement LearningTextBenchmark

🎯 What it does: Propose a quantile-based advantage estimation (QAE) that replaces the traditional mean baseline with a quantile baseline to achieve stable reinforcement learning training for large language model inference tasks.

QuantSparse: Comprehensively Compressing Video Diffusion Transformer with Model Quantization and Attention Sparsification

Weilun Feng (Chinese Academy of Sciences), Yongjun Xu (Chinese Academy of Sciences)

CodeGenerationData SynthesisComputational EfficiencyTransformerDiffusion modelVideo

🎯 What it does: Proposes the QuantSparse framework, which jointly compresses video diffusion Transformers by combining model quantization with sparse attention, achieving significant storage compression and inference acceleration while maintaining or even enhancing generation quality.

Query-Guided Spatial–Temporal–Frequency Interaction for Music Audio–Visual Question Answering

Kun Li (University of Twente), Sami Sebastian Brandt (IT University of Copenhagen)

CodeTransformerPrompt EngineeringVision Language ModelVideoTextMultimodalityBenchmarkAudio

🎯 What it does: For the music audio-visual question answering task, the QSTar method is proposed, which utilizes query-guided multimodal alignment, spatial-temporal-frequency domain interaction, and prompt-based query context reasoning to perform full-process alignment of audio, video, and text features and predict answers.

QueryStream: Advancing Streaming Video Understanding with Query-Aware Pruning and Proactive Response

Kairui Zhang (ShanghaiTech University), Changsheng Xu (ShanghaiTech University)

CodeComputational EfficiencyTransformerVision Language ModelVideoBenchmark

🎯 What it does: Proposes the QueryStream framework, integrating Query-Aware Differential Pruning (QDP) and Relevance-Triggered Active Response (RTAR), for real-time streaming video understanding, significantly reducing the number of visual tokens while improving response accuracy.

QuestA: Expanding Reasoning Capacity in LLMs via Question Augmentation

Jiazheng Li (Tsinghua University), Jingzhao Zhang (Tsinghua University)

CodeOptimizationComputational EfficiencyData-Centric LearningReinforcement LearningPrompt EngineeringTextChain-of-Thought

🎯 What it does: Propose a technique called QUESTA that helps models reason better by inserting partial solutions (hints) for difficult problems during the reinforcement learning training process.

QVGen: Pushing the Limit of Quantized Video Generative Models

Yushi Huang (Hong Kong University of Science and Technology), Jun Zhang (Hong Kong University of Science and Technology)

CodeGenerationComputational EfficiencyKnowledge DistillationDiffusion modelVideo

🎯 What it does: Propose a low-bit quantization-aware training framework (QVGen) for video diffusion models, achieving high-quality video generation under 3-bit/4-bit quantization through auxiliary modules and rank decay;