| PixCon: Clean-Positive Contrastive Learning for Foundation-Model Semi-Supervised Segmentation | Vision & multimodal | Ebenezer Tarubinga | Jul 3, 2026 | 3 |
| RaysUp: Ultra-light Universal Feature Upsampling via Geometry-Aware Ray Representation | Vision & multimodal | Yuchuan Ding +3 | Jun 22, 2026 | 3 |
| How Much Static Structure Do Code Agents Need? A Study of Deterministic Anchoring | Agents | Zhihao Lin +3 | Jun 25, 2026 | 3 |
| The Galaxy's Guide to the Tokenizer: A Benchmark for Scientific Foundation Models | Safety & alignment | Sogol Sanjaripour +5 | Jun 24, 2026 | 3 |
| EO-WM: A Physically Informed World Model for Probabilistic Earth Observation Forecasting | Image & video gen | Junwei Luo +5 | Jun 25, 2026 | 3 |
| How Post-Training Shapes Biological Reasoning Models | Vision & multimodal | Lukas Fesser +7 | Jun 15, 2026 | 3 |
| Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints | Agents | Fangzheng Li +2 | Jun 24, 2026 | 3 |
| Lite Any Stereo V2: Faster and Stronger Efficient Zero-Shot Stereo Matching | Code | Junpeng Jing +7 | Jun 23, 2026 | 3 |
| PrivacyAlign: Contextual Privacy Alignment for LLM Agents | Agents | Manveer Singh Tamber +5 | Jun 19, 2026 | 3 |
| GridVQA-X: A Framework for Evaluating Multimodal Explainability Methods | Vision & multimodal | Sujay Belsare +4 | Jun 2, 2026 | 3 |
| ChartWalker: Benchmarking the Cross-Chart RAG Task | Agents | Ning Tang +8 | Jun 22, 2026 | 3 |
| QG-MIL: A Gated Transformer Aggregator for Domain-Agnostic Multiple Instance Learning in Medical Imaging | Vision & multimodal | Luca Zedda +6 | Jun 18, 2026 | 3 |
| MeshFlow: Mesh Generation with Equivariant Flow Matching | Image & video gen | Qi Sun +8 | Jun 22, 2026 | 3 |
| Arbor: Explicit Geometric Conditioning for Controllable 3D Asset Generation | Efficiency & systems | Jan-Niklas Dihlmann +4 | Jun 22, 2026 | 3 |
| UniverSat: Resolution- and Modality-Agnostic Transformers for Earth Observation | Vision & multimodal | Yohann Perron +4 | Jun 22, 2026 | 3 |
| A Verifiable Search Is Not a Learnable Chain-of-Thought | Efficiency & systems | Harsh Patel | Jun 20, 2026 | 3 |
| Manifold Bandits: Bayesian Curriculum Learning over the Latent Geometry of Large Language Models | Reinforcement learning | Darrien McKenzie +2 | Jun 18, 2026 | 3 |
| Go-with-the-Track: Video Compositing and Motion Control with Point Tracking | Image & video gen | Koichi Namekata +11 | Jun 18, 2026 | 3 |
| Robusto-2: Benchmarking Humans & VLMs for Autonomous Driving in Lima & New York City | Vision & multimodal | Adrian Cespedes +5 | Jun 18, 2026 | 3 |
| Comparing Linear Probes with Mahalanobis Cosine Similarity | Safety & alignment | Zhuofan Josh Ying +2 | Jun 17, 2026 | 3 |
| FastMix: Fast Data Mixture Optimization via Gradient Descent | Code | Haoru Tan +7 | Jun 12, 2026 | 3 |
| StylisticBias: A Few Human Visual Cues Drive Most Social Biases in MLLMs | Vision & multimodal | Shaghayegh Kolli +4 | Jun 18, 2026 | 3 |
| When, Where, and How: Adaptive Binning for Tabular Self-Supervised Learning | Vision & multimodal | Daehwan Kim +2 | Jun 18, 2026 | 3 |
| Characterizing Narrative Content in Web-scale LLM Pretraining Data | Safety & alignment | Teagan Johnson +3 | Jun 17, 2026 | 3 |
| GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning | Robotics | Haoyu Wang +5 | Jun 16, 2026 | 3 |
| SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstruction | Safety & alignment | Yiran Wang +4 | Jun 14, 2026 | 3 |
| JAMER: Project-Level Code Framework Dataset and Benchmark on Professional Game Engines | Agents | Jianwen Sun +7 | Jun 18, 2026 | 3 |
| No Resource, No Benchmarks, No Problem? Evaluating and Improving LLMs for Code Generation in No-Resource Languages | Code | Alessandro Giagnorio +2 | Jun 15, 2026 | 3 |
| The Data Manifold under the Microscope | Data & benchmarks | Marios Koulakis +1 | Jun 14, 2026 | 3 |
| Duration Aware Scheduling for ASR Serving Under Workload Drift | Speech & audio | Darshan Makwana +3 | Mar 11, 2026 | 3 |
| REVES: REvision and VErification--Augmented Training for Test-Time Scaling | Code | Yuanxin Liu +8 | Jun 17, 2026 | 3 |
| Learning User Simulators with Turing Rewards | Agents | Yingshan Susan Wang +7 | Jun 17, 2026 | 3 |
| Reinforcement Learning-Guided Retrieval with Soft Fusion for Robust Multimodal Imitation Learning under Missing Modalities | Robotics | Hassan Ismkhan +1 | Jun 13, 2026 | 3 |
| LLM-Enabled NWDAF: A Step Toward AI-Native 6G Network Intelligence | Code | Henok Daniel +4 | Jun 10, 2026 | 3 |
| iOSWorld: A Benchmark for Personally Intelligent Phone Agents | Vision & multimodal | Lawrence Keunho Jang +5 | Jun 8, 2026 | 3 |
| Beyond Scalar Distances: Semantic Attribute Gradients from Frozen MLLMs for Visual Embeddings | Vision & multimodal | Shubhang Bhatnagar +2 | Jun 13, 2026 | 3 |
| Implicit Reasoning for Large Language Model-based Generative Recommendation | Safety & alignment | Yinhan He +5 | Jun 15, 2026 | 3 |
| AdaSR: Adaptive Streaming Reasoning with Hierarchical Relative Policy Optimization | Speech & audio | Junlong Tong +6 | Jun 12, 2026 | 3 |
| Statistically Reliable LLM-Based Ranking Evaluation via Prediction-Powered Inference | Safety & alignment | Abhishek Divekar | Jun 3, 2026 | 3 |
| Steady-Forcing: Balancing Spatial Persistence and Motion Continuity in Long-Horizon Nature Video Diffusion | Image & video gen | Matiur Rahman Minar +3 | Jun 2, 2026 | 3 |
| Two-Fidelity Best-Action Identification for Stochastic Minimax Tree | Safety & alignment | Peter Chen +1 | Jun 1, 2026 | 3 |
| See What I See, Know What I Think: Dense Latent Communication Across Heterogeneous Agents | Agents | Siyi Chen +9 | Jun 11, 2026 | 3 |
| IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder | Image & video gen | Yitong Chen +7 | Jun 9, 2026 | 3 |
| Revisiting Articulated Parts Perception in Robot Manipulation | Robotics | Xiaoqian Wu +5 | Jun 6, 2026 | 3 |
| Fine-tuning Multi-modal LLMs with ART: Art-based Reinforcement Training | Vision & multimodal | Michal Chudoba +3 | Jun 10, 2026 | 3 |
| Which Models Are Our Models Built On? Auditing Invisible Dependencies in Modern LLMs | Agents | Sanjay Adhikesaven +2 | Jun 10, 2026 | 3 |
| Time-Series Foundation Model Embeddings for Remaining Useful Life Estimation | Data & benchmarks | Amir El-Ghoussani +3 | Jun 10, 2026 | 3 |
| SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference | Code | Yaosheng Fu +4 | Jun 3, 2026 | 3 |
| Large Language Models Are Overconfident in Their Own Responses | Safety & alignment | Mario Sanz-Guerrero +2 | Jun 2, 2026 | 3 |
| Distilling LLM Feedback for Lean Theorem Proving | Reinforcement learning | Gaetan Narozniak +4 | May 29, 2026 | 3 |
| On the Limits of LLM-as-Judge for Scientific Novelty Assessment | Data & benchmarks | Soumitra Sinhahajari +2 | Jun 10, 2026 | 3 |
| The Role of Feedback Alignment in Self-Distillation | Safety & alignment | Semih Kara +1 | Jun 9, 2026 | 3 |
| Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning | Robotics | Zhiyuan Zhou +6 | Jun 9, 2026 | 3 |
| Decentralized Multi-Agent Systems with Shared Context | Agents | Yuzhen Mao +1 | Jun 9, 2026 | 3 |
| Late-Layer Fusion is Enough: Dual-Path Vision Token Routing for Multimodal Large Language Models under Visual Saturation | Image & video gen | Siyuan Liu +1 | Jun 8, 2026 | 3 |
| Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory | Agents | Haoran Sun +10 | Jun 8, 2026 | 3 |
| Phase Marginalization for Patch-Grid Instability in Vision Transformers | Vision & multimodal | Oğuzhan Ercan | Jun 6, 2026 | 3 |
| VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation | Robotics | Siyi Chen +11 | Jun 5, 2026 | 3 |
| SDR: Set-Distance Rewards for Radiology Report Generation | Vision & multimodal | Halil Ibrahim Gulluk +3 | May 30, 2026 | 3 |
| WorldBench: A Challenging and Visually Diverse Multimodal Reasoning Benchmark | Vision & multimodal | Yida Yin +11 | Jun 4, 2026 | 3 |
| Reinforcement Learning from Rich Feedback with Distributional DAgger | Code | Rishabh Agrawal +2 | Jun 3, 2026 | 3 |
| When Gradients Collide: Failure Modes of Multi-Objective Prompt Optimization for LLM Judges | Efficiency & systems | Parth Darshan +1 | May 25, 2026 | 3 |
| Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models | Vision & multimodal | Haibo Wang +1 | Jun 4, 2026 | 3 |
| BRepCLIP: Contrastive Multimodal Pretraining on BRep Primitives for CAD Understanding | Vision & multimodal | Muhammad Usama +3 | Jun 3, 2026 | 3 |
| Video2LoRA: Parametric Video Internalization for Vision-Language Models | Vision & multimodal | Manan Suri +2 | Jun 3, 2026 | 3 |
| EvoDS: Self-Evolving Autonomous Data Science Agent with Skill Learning and Context Management | Agents | Zherui Yang +3 | Jun 2, 2026 | 3 |
| Trust Region Q Adjoint Matching | Reinforcement learning | Yonghoon Dong +4 | May 26, 2026 | 3 |
| STRIDE: Training Data Attribution via Sparse Recovery from Subset Perturbations | Efficiency & systems | Rishit Dagli +6 | Jun 3, 2026 | 3 |
| Evaluating Large Language Models in Dynamic Clinical Decision-Making with Standardized Patient Cases | Agents | Cheng Liang +5 | Jun 3, 2026 | 3 |
| Agent libOS: A Library-OS-Inspired Runtime for Long-Running, Capability-Controlled LLM Agents | Agents | Yingqi Zhang | Jun 2, 2026 | 3 |
| SpatialAct: Probing Spatial Reasoning-to-Action Capabilities of VLM Agents in 3D Scenes | Vision & multimodal | Tianhui Liu +8 | May 29, 2026 | 3 |
| OpenSTBench: Beyond Semantic Evaluation for Speech Translation | Speech & audio | Yanjie An +7 | May 29, 2026 | 3 |
| PaintBench: Deterministic Evaluation of Precise Visual Editing | Robotics | Kai Xu +5 | May 29, 2026 | 3 |
| The Meta-Agent Challenge: Are Current Agents Capable of Autonomous Agent Development? | Agents | Xinyu Lu +10 | Jun 3, 2026 | 3 |
| AURA: Action-Gated Memory for Robot Policies at Constant VRAM | Robotics | Josef Chen | Jun 1, 2026 | 3 |
| Prior Availability in Industrial Visual Sim-to-Real: A Review of CAD-Guided and CAD-Unavailable Regimes | Vision & multimodal | Chenxi Tao +1 | May 28, 2026 | 3 |
| EVA01: Unified Native 3D Understanding and Generation via Mixture-of-Transformers | Image & video gen | Zongyuan Yang +10 | May 16, 2026 | 3 |
| When Confidence Misleads: Suffix Anchoring and Anchor-Proximity Confidence Modulation for Diffusion Language Models | Image & video gen | Jungwon Park +4 | May 27, 2026 | 3 |
| Flat-Pack Bench: Evaluating Spatio-Temporal Understanding in Large Vision-Language Models through Furniture Assembly | Vision & multimodal | Aditya Chetan +7 | May 20, 2026 | 3 |
| OmniInteract: Benchmarking Real-World Streaming Interaction for Real-Time Omnimodal Assistants | Speech & audio | Xudong Lu +10 | May 26, 2026 | 3 |
| Chartographer: Counterfactual Chart Generation for Evaluating Vision-Language Models | Vision & multimodal | Yifan Jiang +3 | May 26, 2026 | 3 |
| Cross-scale Aligned Supervision for Training GANs | Image & video gen | Sangeek Hyun +2 | May 26, 2026 | 3 |
| Seeing the Needle in the Haystack: Towards Weakly-Supervised Log Instance Anomaly Localization via Counterfactual Perturbation | Vision & multimodal | Yutszyuk Wong +3 | May 9, 2026 | 3 |
| Self-Guided Test-Time Training for Long-Context LLMs | Data & benchmarks | Xinyu Zhu +12 | Jul 10, 2026 | 2 |
| Token-Based Dual-view Fusion and Adaptation of Large Vision Models for Breast Cancer Classification | Vision & multimodal | Aysan Ghayouri Pirsoltan +2 | Jul 7, 2026 | 2 |
| Imagined Rollouts are Kinematic, Not Dynamic: A Diagnosis of Long-Horizon World-Model Failure | Image & video gen | Finn Rasmus Schäfer +5 | Jul 7, 2026 | 2 |
| TESSERA v2: Scaling Pixel-wise Earth Foundation Models | Code | Zhengpeng Feng +12 | Jul 4, 2026 | 2 |
| Wake up for Touch! Mask-isolated Tactile Alignment Learning in MLLMs | Vision & multimodal | Yoonhyung Park +3 | Jul 1, 2026 | 2 |
| Image2Sim: Scaling Embodied Navigation via Generative Neural Simulator | Robotics | Zihan Wang +3 | Jul 7, 2026 | 2 |
| Boundary-Aware Context Grounding for A Low-Channel EEG Agent | Agents | Zhiyuan Xu +3 | Jun 25, 2026 | 2 |
| Physics Question Scene Graph: Fine-grained Evaluation of Physical Plausibility in Text-to-Video Generation | Image & video gen | Atin Pothiraj +4 | Jun 24, 2026 | 2 |
| Distill Once, Adapt Life-Long: Exploring Dataset Distillation for Continual Test-Time Adaptation | Code | Hyun-Kurl Jang +3 | Jun 18, 2026 | 2 |
| When Agents Commit Too Soon: Diagnosing Premature Commitment in LLM Agents | Agents | Aman Mehta | Jun 22, 2026 | 2 |
| Capable but Careless: Do Computer-Use Agents Follow Contextual Integrity? | Vision & multimodal | Anmol Goel +1 | Jun 22, 2026 | 2 |
| Libretto: Giving LLM Agents a Sense of Musical Structure | Speech & audio | Yichen Xu | Jun 21, 2026 | 2 |
| ReSyn: A Generalized Recursive Regular Expression Synthesis Framework | Code | Seongmin Kim +4 | Jun 13, 2026 | 2 |
| Bag of Dims: Training-Free Mechanistic Interpretability via Dimension-Level Sign Patterns | Speech & audio | Varun Reddy Nalagatla | Jun 17, 2026 | 2 |
| Seeing Before Reasoning: Decoupling Perception and Reasoning for Shortcut-Resilient Multimodal On-Policy Self-Distillation | Vision & multimodal | Sihan Wang +3 | Jun 17, 2026 | 2 |
| EgoPhys: Learning Generalizable Physics Models of Deformable Objects from Egocentric Video | Robotics | Hyunjin Kim +3 | Jun 15, 2026 | 2 |
| Selective Control under Noisy Perception: Governance Failures Hidden by Aggregate Metrics in Modular Networks | Agents | Igor Itkin | Jun 12, 2026 | 2 |