Monthly collection

June 2026

The papers that drew the most attention during this month of AI research.

Papers: 86
Citations: 164
Research labs: 5

Explore the year

View all months →

Aug2025 Sep2025 Oct2025 Nov2025 Dec2025 Jan2026 Feb2026 Mar2026 Apr2026 May2026 Jun2026 Jul2026 Aug2026

1,016papers
curated and indexed

86 papers published in June 2026

01

Research paper

ABot-Earth 0.5: Generative 3D Earth Model

ABot-Earth 0.5, developed by AMAP CV Lab (Alibaba Group), is a generative 3D framework that synthesizes vast, seamless 3D environments from geospatially referenced satellite imagery using a native 3D Gaussian Splatting (3DGS) representation. Trained on real-world urban reconstructions, it generates realistic geometry and textures at under 10 minutes per…

Ming Qian, Tianjian Ouyang, Mingchao Sun, Zijian Wang, et al.

Published: Jun 2026
Upvotes: 488
Citations: 1

Read summary Original

02

Research paper

Looped World Models

The paper introduces Looped World Models (LoopWM), the first looped transformer architecture for world modeling, addressing the tension between deep computation for faithful long-horizon simulation and the high cost and error accumulation of deep models. LoopWM iteratively refines latent environment states through a parameter-shared transformer block with…

Hongyuan Adam Lu, Z. L. Victor Wei, Qun Zhang, Jinrui Zeng, et al.

Published: Jun 2026
Upvotes: 484
Citations: 1

Read summary Original

03

Research paper

Orca: The World is in Your Mind

Orca, developed by the Beijing Academy of Artificial Intelligence, is a general world foundation model that learns a unified world latent space from multimodal signals (vision and language) using Next-State-Prediction modeling. It employs two complementary learning paradigms: unconscious learning captures dense natural state transitions from continuous…

Yihao Wang, Yuheng Ji, Mingyu Cao, Yanqing Shen, et al.

Published: Jun 2026
Upvotes: 474
Citations: 0

Read summary Original

04

Research paper

Agents' Last Exam

Agents' Last Exam (ALE) is a benchmark introduced by UC Berkeley and collaborators to evaluate AI agents on long-horizon, economically valuable, real-world professional tasks with verifiable outcomes. Developed with 250+ industry experts, ALE covers 55 subfields across 13 industry clusters, grounded in the O*NET/SOC 2018 occupational taxonomy, and includes…

Yiyou Sun, Xinyang Han, Weichen Zhang, Yuanbo Pang, et al.

Published: Jun 2026
Upvotes: 387
Citations: 5

Read summary Original

05

Research paper

On the Scaling of PEFT: Towards Million Personal Models of Trillion Parameters

This paper argues that parameter-efficient fine-tuning (PEFT), especially LoRA, can scale from a single shared foundation model to millions of persistent personal model instances. The authors propose a three-axis framework: Scale Up (strengthening the shared base model), Scale Down (shrinking the local adaptive state), and Scale Out (sustaining a large…

Mind Lab, :, Vin Bo, Song Cao, et al.

Published: Jun 2026
Upvotes: 241
Citations: 2

Read summary Original

06

Research paper

JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence

JoyAI-VL-Interaction introduces a paradigm shift from turn-based to proactive, streaming interaction for vision-language models. The authors release an 8B-scale, vision-first model that continuously watches live video and decides each second to stay silent, respond, or delegate complex tasks to a background model. Built on JoyAI-VL 1.0 with AdaCodec for…

Dingyu Yao, Junhao Zhou, Chenxu Yang, Chuanyu Qin, et al.

Published: Jun 2026
Upvotes: 216
Citations: 2

Read summary Original

07

Research paper

LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling

The paper investigates loop-count selection in Parallel Loop Transformers (PLT), which use cross-loop position offsets (CLP) and shared-KV gated sliding-window attention to keep latency and memory constant regardless of loop count. The authors train LoopCoder-v2, a 7B PLT coder, from scratch on 18T tokens with loop counts R=1,2,3,4, under matched training…

Jian Yang, Shawn Guo, Wei Zhang, Tianyu Zheng, et al.

Published: Jun 2026
Upvotes: 210
Citations: 1

Read summary Original

08

Research paper

Kwai Keye-VL-2.0 Technical Report

The report introduces Kwai Keye-VL-2.0-30B-A3B, an open-source Mixture-of-Experts (MoE) multimodal foundation model with 30B total parameters and 3B active, designed for long-video understanding and agentic intelligence. It is the first to adapt DeepSeek Sparse Attention (DSA) to GQA-based architectures, enabling lossless 256K context processing. The model…

Kwai Keye Team, Bin Wen, Changyi Liu, Chengru Song, et al.

Published: Jun 2026
Upvotes: 194
Citations: 0

Read summary Original

09

Research paper

MemSlides: A Hierarchical Memory Driven Agent Framework for Personalized Slide Generation with Multi-turn Local Revision

MemSlides is a hierarchical memory framework for personalized presentation generation that separates long-term memory (user profile memory and tool memory) from working memory. User profile memory stores intent-conditioned preferences for round-0 personalization, working memory carries active preferences and session constraints across revision rounds, and…

Ye Jin, Yangyang Xu, Jun Zhu, Yibo Yang

Published: Jun 2026
Upvotes: 177
Citations: 0

Read summary Original

10

Research paper

Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models

Embodied-R1.5 is an 8B-parameter Embodied Foundation Model (EFM) that unifies three embodied reasoning dimensions—cognition and spatial reasoning, planning and correction, and pointing and location—within a single architecture. It uses three automated data pipelines to build a 15B-token corpus and a multi-task balanced RL recipe to resolve heterogeneous…

Yifu Yuan, Yaoting Huang, Xianze Yao, Yutong Li, et al.

Published: Jun 2026
Upvotes: 172
Citations: 1

Read summary Original

11

Research paper

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

The paper addresses training-inference mismatch in LLM reinforcement learning, where separate training and inference engines produce inconsistent probabilities for the same trajectories, causing off-policyness and training instability. The authors identify an objective misalignment: improving the training policy does not guarantee improvement of the…

Jing Liang, Hongyao Tang, Yi Ma, Yancheng He, et al.

Published: Jun 2026
Upvotes: 170
Citations: 0

Read summary Original

12

MiniMax

MiniMax Sparse Attention

MiniMax Sparse Attention (MSA) is a blockwise sparse attention mechanism built on Grouped Query Attention (GQA) to address the quadratic cost of softmax attention in ultra-long contexts. A lightweight Index Branch scores key-value blocks and selects a Top-k subset per GQA group, while the Main Branch computes exact block-sparse attention over selected…

Xunhao Lai, Weiqi Xu, Yufeng Yang, Qiaorui Chen, et al.

Published: Jun 2026
Upvotes: 154
Citations: 7

Read summary Original

13

Qwen

Qwen-AgentWorld: Language World Models for General Agents

Qwen-AgentWorld introduces the first language world models (LWMs) for simulating agentic environments across seven domains (MCP, Search, Terminal, SWE, Android, Web, OS). Trained on over 10 million real-world interaction trajectories via a three-stage pipeline (CPT, SFT, RL), the models (35B-A3B and 397B-A17B) predict next environment states given actions.…

Yuxin Zuo, Zikai Xiao, Li Sheng, Fei Huang, et al.

Published: Jun 2026
Upvotes: 154
Citations: 0

Read summary Original

14

Research paper

Agentic Abstention: Do Agents Know When to Stop Instead of Act?

The paper introduces Agentic Abstention, the problem of deciding when an LLM agent should stop acting and abstain rather than continue interacting with an environment. Unlike single-turn LLM abstention, this is a sequential decision problem where agents can answer, abstain, or act at each turn. The authors construct a benchmark of over 28,000 tasks across…

Han Luo, Bingbing Wen, Lucy Lu Wang

Published: Jun 2026
Upvotes: 150
Citations: 0

Read summary Original

15

Research paper

EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments

The paper introduces EvoArena, a benchmark suite for evaluating LLM agents under persistent environment evolution, and EvoMem, a patch-based memory paradigm that records memory updates as structured histories. EvoArena comprises three subsets: Terminal-Bench-Evo (evolving terminal workflows), SWE-Chain-Evo (evolving codebases), and PersonaMem-Evo (evolving…

Jundong Xu, Qingchuan Li, Jiaying Wu, Yihuai Lan, et al.

Published: Jun 2026
Upvotes: 143
Citations: 0

Read summary Original

16

NVIDIA

Cosmos 3: Omnimodal World Models for Physical AI

NVIDIA introduces Cosmos 3, a family of omnimodal world models that jointly process and generate language, image, video, audio, and action sequences within a unified mixture-of-transformers architecture. It subsumes vision-language models, video generators, world simulators, and world-action models into a single framework, supporting flexible input-output…

NVIDIA, :, Aditi, Niket Agarwal, et al.

Published: Jun 2026
Upvotes: 142
Citations: 36

Read summary Original

17

Research paper

RESOURCE2SKILL: Distilling Executable Agent Skills from Human-Created Multimodal Resources

Resource2Skill is a framework that automatically distills multimodal human-created resources—tutorial videos, repositories, articles, and reference artifacts—into executable skills for software agents, organized as a hierarchical multimodal Skill Wiki. Each skill entry combines structured text, code, visual examples, metadata, and provenance, preserving…

Yijia Fan, Zonglin Di, Zimo Wen, Yifan Yang, et al.

Published: Jun 2026
Upvotes: 142
Citations: 0

Read summary Original

18

Research paper

Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

Moebius is a highly efficient lightweight image inpainting framework that rivals the generation quality of 10B-level industrial models like FLUX.1-Fill-Dev while using only 0.22B parameters (less than 2% of FLUX's 11.9B) and delivering over 15× faster total inference time. To overcome the representation bottleneck from extreme structural compression, the…

Kangsheng Duan, Ziyang Xu, Wenyu Liu, Xiaohu Ruan, et al.

Published: Jun 2026
Upvotes: 141
Citations: 0

Read summary Original

19

Research paper

Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories

The paper introduces Data Journalist Agent (Data2Story), a multi-agent framework that transforms raw data into verifiable, multimodal articles. It orchestrates seven roles (Detective, Analyst, Editor, Designer, Programmer, Auditor, Inspector) into a virtual newsroom. Key innovations are evidence-grounded claims via an Inspector that links each claim to…

Kevin Qinghong Lin, Batu EI, Yuhong Shi, Pan Lu, et al.

Published: Jun 2026
Upvotes: 132
Citations: 0

Read summary Original

20

Research paper

Are We Ready For An Agent-Native Memory System?

This paper presents a systematic experimental study of agent memory systems for LLM agents, decomposing them into four core modules: representation/storage, extraction, retrieval/routing, and maintenance. The authors evaluate 12 representative memory systems and two baselines across five benchmark workloads (11 datasets), finding that no single…

Wei Zhou, Xuanhe Zhou, Shaokun Han, Hongming Xu, et al.

Published: Jun 2026
Upvotes: 131
Citations: 7

Read summary Original

21

Research paper

Toward Generalist Autonomous Research via Hypothesis-Tree Refinement

The paper introduces Arbor, a general framework for autonomous research under the Autonomous Optimization (AO) setting, where an agent iteratively improves an initial artifact without step-level human supervision. Arbor uses Hypothesis Tree Refinement (HTR), combining a long-lived coordinator that maintains a persistent hypothesis tree (linking hypotheses,…

Jiajie Jin, Yuyang Hu, Kai Qiu, Qi Dai, et al.

Published: Jun 2026
Upvotes: 130
Citations: 2

Read summary Original

22

Research paper

Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models

The paper introduces Imaginative Perception Tokens (IPTs), intermediate visual representations that externalize what a VLM would perceive under an alternative spatial configuration, to improve spatial reasoning. Three tasks requiring imaginative perception are formulated: Perspective Taking (PET), Path Tracing (PT), and Multiview Counting (MVC), with…

Mahtab Bigverdi, Linjie Li, Weikai Huang, Yiming Liu, et al.

Published: Jun 2026
Upvotes: 126
Citations: 0

Read summary Original

23

Research paper

VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models

VibeThinker-3B is a 3B-parameter dense model from Sina Weibo Inc. that achieves frontier-level performance on verifiable reasoning tasks, matching or exceeding much larger models. Built on Qwen2.5-Coder-3B, it uses a post-training pipeline with curriculum SFT, multi-domain RL (math, code, STEM), Long2Short Math RL for efficiency, offline self-distillation,…

Sen Xu, Shixi Liu, Wei Wang, Jixin Min, et al.

Published: Jun 2026
Upvotes: 125
Citations: 1

Read summary Original

24

Research paper

SWE-Explore: Benchmarking How Coding Agents Explore Repositories

SWE-Explore is a benchmark that isolates and evaluates the repository exploration capability of coding agents, independent of patch generation. It formalizes exploration as a ranked, line-level context selection task: given an issue and repository, an explorer returns a ranked list of code regions under a fixed line budget. The benchmark includes 848…

Shaoqiu Zhang, Yuhang Wang, Jialiang Liang, Yuling Shi, et al.

Published: Jun 2026
Upvotes: 123
Citations: 10

Read summary Original

25

Research paper

Audio Interaction Model

The paper introduces Audio-Interaction, a unified streaming audio language model that operates via an always-on perceive–decide–respond loop, listening to continuous audio and deciding when to respond or remain silent. It addresses limitations of offline LALMs and task-specific streaming models by unifying capabilities like real-time ASR, translation,…

Zhifei Xie, Zihang Liu, Ze An, Xiaobin Hu, et al.

Published: Jun 2026
Upvotes: 121
Citations: 0

Read summary Original

26

Research paper

Wan-Streamer v0.1: End-to-end Real-time Interactive Foundation Models

Wan-Streamer is a native-streaming, end-to-end interactive foundation model from Alibaba Group designed for real-time, low-latency, full-duplex audio-visual interaction. It models language, audio, and video as both input and output within a single Transformer, using block-causal attention for incremental streaming. Unlike cascaded systems, it does not rely…

Lianghua Huang, Zhi-Fan Wu, Wei Wang, Yupeng Shi, et al.

Published: Jun 2026
Upvotes: 120
Citations: 4

Read summary Original

27

Research paper

Geometric Action Model for Robot Policy Learning

The paper introduces the Geometric Action Model (GAM), a language-conditioned manipulation policy that repurposes a pretrained geometric foundation model (GFM) as a shared substrate for perception, temporal prediction, and action decoding. GAM splits the GFM at an intermediate layer: shallow layers encode observations, a causal future predictor inserted at…

Jisang Han, Seonghu Jeon, Jaewoo Jung, René Zurbrügg, et al.

Published: Jun 2026
Upvotes: 119
Citations: 1

Read summary Original

28

Research paper

From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI

This survey paper by Tencent Youtu Lab and academic partners frames the evolution of Large Language Models (LLMs) as a shift from conversational chatbots to persistent, autonomous 'Digital Colleagues.' The authors organize this transition along two dimensions: the cognitive core (from fast, next-token-prediction 'Chatbot' models to 'Thinking LLMs' that use…

Yongheng Zhang, Ziang Liu, Jiaxuan Zhu, Shuai Wang, et al.

Published: Jun 2026
Upvotes: 117
Citations: 3

Read summary Original

29

Research paper

DreamX-World 1.0: A General-Purpose Interactive World Model

DreamX-World 1.0 is a general-purpose interactive text/image-to-video world model for controllable long-horizon generation, supporting camera navigation, revisits, and promptable events across photorealistic, game-style, and stylized domains. It uses a data engine combining Unreal Engine rendering, gameplay recordings, and real-world videos. The model…

DreamX Team, Yancheng Bai, Rui Chen, Xiangxiang Chu, et al.

Published: Jun 2026
Upvotes: 114
Citations: 5

Read summary Original

30

Research paper

Dockerless: Environment-Free Program Verifier for Coding Agents

Dockerless is an environment-free agentic program verifier that evaluates code patches without executing them, using agentic repository exploration to gather evidence. It generates verification questions from the issue and reference patch, dispatches parallel sub-agents to collect evidence-backed answers, and produces a correctness score. Trained via…

Wenhao Zeng, Yuling Shi, Xiaodong Gu, Chao Hu, et al.

Published: Jun 2026
Upvotes: 114
Citations: 1

Read summary Original

31

Research paper

OmniDirector: General Multi-Shot Camera Cloning without Cross-Paired Data

OmniDirector is a framework for cloning camera motion from reference videos to animate source images, supporting multi-shot sequences without requiring cross-paired training data. It introduces a 'camera grid' representation, which renders camera parameters as a grid motion video within an empty 3D scene, decoupling camera motion from content and enabling…

Jiwen Liu, Shujuan Li, Zhixue Fang, Xiaohan Li, et al.

Published: Jun 2026
Upvotes: 113
Citations: 5

Read summary Original

32

Research paper

DOPD: Dual On-policy Distillation

The paper introduces DOPD (Dual On-policy Distillation), an advantage-aware dual distillation paradigm for large language models (LLMs) and vision-language models (VLMs). It addresses a failure mode called 'privilege illusion,' where adding privileged information to teacher or student policies creates an apparent performance gap that stems from information…

Xinlei Yu, Gen Li, Qingyi Si, Guibin Zhang, et al.

Published: Jun 2026
Upvotes: 113
Citations: 1

Read summary Original

33

NVIDIA

SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning

SpatialClaw is a training-free framework that improves spatial reasoning in vision-language models (VLMs) by using code as the action interface. It maintains a persistent Python kernel pre-loaded with input frames and perception tools, allowing a VLM-backed agent to write and execute one code cell per step, inspect intermediate results (e.g., masks, depth…

Seokju Cho, Ryo Hachiuma, Abhishek Badki, Hang Su, et al.

Published: Jun 2026
Upvotes: 111
Citations: 3

Read summary Original

34

Research paper

WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces

WeaveBench is a long-horizon, hybrid-interface benchmark for computer-use agents, comprising 114 tasks across 8 real-world work domains. Each task requires agents to combine GUI observations/actions with CLI/code operations within a single trajectory, satisfying three admission criteria: channel non-substitutability, long-horizon execution, and…

Wanli Li, Bowen Zhou, Yunyao Yu, Zhou Xu, et al.

Published: Jun 2026
Upvotes: 107
Citations: 2

Read summary Original

35

Research paper

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

Agents-A1 is a 35B Mixture-of-Experts agentic model that achieves trillion-parameter-level performance by scaling the agent horizon rather than parameters. The authors built a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories averaging 45K tokens.…

Lei Bai, Zongsheng Cao, Yang Chen, Zhiyao Cui, et al.

Published: Jun 2026
Upvotes: 103
Citations: 3

Read summary Original

36

Research paper

Your UnEmbedding Matrix is Secretly a Feature Lens for Text Embeddings

The paper identifies a cause of LLMs' suboptimal zero-shot text embedding performance: text embeddings align with high-frequency but uninformative tokens when projected onto the vocabulary space. Using Logit Lens and Logit Spectroscopy, the authors discover an 'edge spectrum' subspace in the unembedding matrix that encodes these frequent tokens. They…

Songhao Wu, Zhongxin Chen, Yuxuan Liu, Heng Cui, et al.

Published: Jun 2026
Upvotes: 99
Citations: 0

Read summary Original

37

Research paper

Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution

Code2LoRA is a hypernetwork framework that generates repository-specific LoRA adapters for code language models, injecting repository knowledge with zero inference-time token overhead. It has two usage scenarios: Code2LoRA-Static maps a single repository snapshot to an adapter for stable codebases, and Code2LoRA-Evo maintains an adapter via a GRU hidden…

Liliana Hotsko, Yinxi Li, Yuntian Deng, Pengyu Nie

Published: Jun 2026
Upvotes: 96
Citations: 1

Read summary Original

38

Research paper

PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems

PlanBench-XL is a new interactive benchmark for evaluating long-horizon planning of LLM tool-use agents in large-scale tool ecosystems. It consists of 327 retail tasks over 1,665 tools, requiring agents to iteratively retrieve tools, infer implicit sub-goals, and adapt to dynamic environments. The benchmark features a retrieval-mediated environment with…

Jiayu Liu, Qihan Lin, Cheng Qian, Rui Wang, et al.

Published: Jun 2026
Upvotes: 96
Citations: 4

Read summary Original

39

MiniMax

MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling

MaxProof is a population-level test-time scaling framework for competition-level mathematical proof, developed for the MiniMax-M3 model series. The M3 model is trained with three proof-oriented capabilities: proof generation via verifier-guided reinforcement learning, proof verification through aligned error finding, and critique-conditioned proof repair.…

Jiacheng Chen, Xinyu Zhang, Shunkai Zhang, Yanmohan Wang, et al.

Published: Jun 2026
Upvotes: 94
Citations: 1

Read summary Original

40

Research paper

Redesign Mixture-of-Experts Routers with Manifold Power Iteration

This paper introduces Manifold Power Iteration (MPI), a redesign of Mixture-of-Experts (MoE) routers to align each router row with the principal singular direction of its associated expert's weight matrix. The method uses a 'Power-then-Retract' paradigm: a single power iteration step on router weights followed by L2 norm retraction to ensure stability.…

Songhao Wu, Ang Lv, Ruobing Xie, Yankai Lin

Published: Jun 2026
Upvotes: 90
Citations: 0

Read summary Original

41

Research paper

Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale

The report introduces Ling-2.6 and Ring-2.6, a family of trillion-parameter models for efficient agentic intelligence. Ling-2.6 is optimized for instant responses and high token efficiency, while Ring-2.6 targets deeper reasoning and advanced agentic workflows. Instead of training from scratch, the models upgrade the Ling-2.0 base via architectural…

Ang Li, Ben Liu, Bin Han, Bin Hu, et al.

Published: Jun 2026
Upvotes: 87
Citations: 2

Read summary Original

42

Research paper

InterleaveThinker: Reinforcing Agentic Interleaved Generation

InterleaveThinker is a multi-agent framework that endows existing image generators with interleaved text-image generation capabilities, addressing visual over-reliance and step-wise error accumulation in Unified Multimodal Models (UMMs). It uses a Planner agent to pre-plan the full instruction sequence, a Generator (e.g., FLUX.2-klein-9B) to execute steps,…

Dian Zheng, Harry Lee, Manyuan Zhang, Kaituo Feng, et al.

Published: Jun 2026
Upvotes: 84
Citations: 0

Read summary Original

43

Research paper

Robust-U1: Can MLLMs Self-Recover Corrupted Visual Content for Robust Understanding?

Multimodal Large Language Models (MLLMs) degrade under real-world visual corruptions. Existing robustness methods are limited: black-box feature alignment lacks interpretability, and text-based reasoning cannot restore pixel-level details. This paper proposes Robust-U1, a framework that equips MLLMs with explicit visual self-recovery capability. It uses a…

Jiaqi Tang, Jianmin Chen, Youyang Zhai, Wei Wei, et al.

Published: Jun 2026
Upvotes: 82
Citations: 1

Read summary Original

44

Research paper

LiveEdit: Towards Real-Time Diffusion-Based Streaming Video Editing

LiveEdit is a novel streaming video editing framework that performs causal, chunk-by-chunk editing with high fidelity and ultra-low latency. It addresses two core issues: attention distribution shift and spatial-temporal token redundancy. The method uses a three-stage distillation pipeline: Stage 1 tunes a bidirectional DiT for editing, Stage 2 transitions…

Xinyu Wang, Chongbo Zhao, Fangneng Zhan, Yue Ma

Published: Jun 2026
Upvotes: 82
Citations: 2

Read summary Original

45

Research paper

DanceOPD: On-Policy Generative Field Distillation

DanceOPD is an on-policy generative field distillation framework for flow-matching image generation models, designed to compose multiple capabilities (e.g., text-to-image, local editing, global editing) into a single student model. The method treats each frozen capability source as a velocity field over a shared state space and addresses three key…

Wei Zhou, Xiongwei Zhu, Zelin Xu, Bo Dong, et al.

Published: Jun 2026
Upvotes: 81
Citations: 2

Read summary Original

46

Research paper

Does VLA Even Know the Basics? Measuring Commonsense and World Knowledge Retention in Vision-Language-Action Models

The paper introduces Act2Answer, a protocol for evaluating commonsense and world knowledge retention in Vision-Language-Action (VLA) models. It converts VLM knowledge benchmarks into embodied tabletop episodes where agents answer by placing a cube on one of two candidate images, reducing control confounds. The suite includes 1,720 binary questions across…

Nikita Kachaev, Andrey Moskalenko, Matvey Skripkin, Nikita Kurlaev, et al.

Published: Jun 2026
Upvotes: 80
Citations: 0

Read summary Original

47

Research paper

Grouped Query Experts: Mixture-of-Experts on GQA Self-Attention

The paper introduces Grouped Query Experts (GQE), a mixture-of-experts layer applied to grouped-query attention (GQA) in Transformers. GQE routes each token to k query-head experts within each GQA group, while keeping all key-value (KV) heads dense and unchanged, preserving GQA's KV-cache benefits and reducing active query-head computation. The method…

Vishesh Tripathi, Abhay Kumar

Published: Jun 2026
Upvotes: 80
Citations: 0

Read summary Original

48

Research paper

EnterpriseClawBench: Benchmarking Agents from Real Workplace Sessions

EnterpriseClawBench is an enterprise agent benchmark built from 5,291 real workplace sessions at an AI startup, yielding 852 reproducible tasks (120 in a manually audited Lite subset). The pipeline applies mechanical gates (length, fixture, redaction, network) and rewrites prompts into single-turn tasks with role/skill taxonomies, hard rules, and semantic…

Jincheng Zhong, Weizhi Wang, Che Jiang, Kai Tian, et al.

Published: Jun 2026
Upvotes: 80
Citations: 0

Read summary Original

49

Research paper

FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents

The paper introduces FORT-Searcher, a deep search agent trained using FORT, a framework for synthesizing shortcut-resistant training data. The authors argue that existing methods for creating deep search tasks often fail to induce real search difficulty because agents can exploit shortcuts, such as evidence co-coverage, single-clue selectivity, exposed…

Jia Deng, Yimeng Chen, Xiaoqing Xiang, Ziyang Zeng, et al.

Published: Jun 2026
Upvotes: 79
Citations: 1

Read summary Original

50

Research paper

APPO: Agentic Procedural Policy Optimization

The paper introduces APPO (Agentic Procedural Policy Optimization), a reinforcement learning algorithm for LLM agents that shifts branching and credit assignment from coarse units like tool-call boundaries to fine-grained decision points in the generated sequence. A pilot study showed that influential decision points are distributed throughout the thinking…

Xucong Wang, Ziyu Ma, Yong Wang, Yuxiang Ji, et al.

Published: Jun 2026
Upvotes: 79
Citations: 0

Read summary Original

51

Research paper

OpenRath: Session-Centered Runtime State for Agent Systems

OpenRath addresses the hidden-runtime-state problem in multi-agent systems by introducing Session, a first-class runtime value that carries conversation chunks, lineage, sandbox placement, tool evidence, memory events, and usage. The programming model, inspired by PyTorch's architecture, defines objects like Agent, Workflow, Tool, Memory, Sandbox, and…

Fukang Wen, Zhijie Wang, Ruilin Xu

Published: Jun 2026
Upvotes: 78
Citations: 0

Read summary Original

52

Research paper

Memory is Reconstructed, Not Retrieved: Graph Memory for LLM Agents

The paper introduces MRAgent, a framework for LLM agents that treats memory access as an active, multi-step reconstruction process over a structured memory graph, rather than a passive retrieve-then-reason pipeline. MRAgent organizes memory as a Cue–Tag–Content graph, where associative tags link fine-grained cues to memory contents, enabling the LLM to…

Shuo Ji, Yibo Li, Bryan Hooi

Published: Jun 2026
Upvotes: 77
Citations: 2

Read summary Original

53

Research paper

Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution

Role-Agent is a framework that enables bootstrapped agent-environment co-evolution using a single LLM in dual roles. It consists of World-In-Agent (WIA), where the LLM predicts future states after actions, using the alignment between predicted and actual states as a process reward to encourage environment-aware reasoning, and Agent-In-World (AIW), where…

Xucong Wang, Ziyu Ma, Shidong Yang, Tongwen Huang, et al.

Published: Jun 2026
Upvotes: 77
Citations: 0

Read summary Original

54

Research paper

Learning from the Self-future: On-policy Self-distillation for dLLMs

This paper introduces d-OPSD, the first on-policy self-distillation (OPSD) framework tailored for diffusion large language models (dLLMs). Existing OPSD methods are autoregressive-centric, using left-to-right prefix conditioning and token-level supervision, which conflicts with dLLMs' arbitrary-order generation. d-OPSD makes two key contributions: it…

Yifu Luo, Zeyu Chen, Haoyu Wang, Xinhao Hu, et al.

Published: Jun 2026
Upvotes: 77
Citations: 1

Read summary Original

55

Research paper

BlockPilot: Instance-Adaptive Policy Learning for Diffusion-based Speculative Decoding

BlockPilot introduces a sample-adaptive policy for diffusion-based speculative decoding, addressing the inefficiency of fixed block sizes. The authors observe that the optimal block size varies across samples but concentrates near the training block size, enabling a lightweight classification approach. BlockPilot predicts the block size from the last…

Hao Zhang, Yiming Hu, Yong Wang, Mingqiao Mo, et al.

Published: Jun 2026
Upvotes: 77
Citations: 0

Read summary Original

56

Research paper

Unlimited OCR Works

Baidu's Unlimited OCR introduces Reference Sliding Window Attention (R-SWA) to enable one-shot long-horizon document parsing. R-SWA replaces all attention layers in the decoder of DeepSeek OCR, allowing each generated token to attend to all reference tokens (visual and prompt) and a causal sliding window of the previous 128 output tokens. This maintains a…

Youyang Yin, Huanhuan Liu, YY, Qunyi Xie, et al.

Published: Jun 2026
Upvotes: 76
Citations: 4

Read summary Original

57

Research paper

On the Geometry of On-Policy Distillation

This paper analyzes the parameter-space geometry of on-policy distillation (OPD) for large language models, comparing it with supervised fine-tuning (SFT) and reinforcement learning with verifiable rewards (RLVR). Using diagnostics like update sparsity, subspace rotation, spectral drift, and update localization, the authors find that OPD occupies a…

Zhennan Shen, Yanshu Li, Qingyu Yin, Chak Tou Leong, et al.

Published: Jun 2026
Upvotes: 75
Citations: 2

Read summary Original

58

Research paper

DataClaw0: Agentic Tailoring Multimodal Data from Raw Streams

DataClaw0 is a framework for intent-conditioned multimodal data tailoring, converting raw, noisy streams (e.g., tutorial videos, embodied trajectories, GUI logs) into schema-aligned, evidence-grounded training instances. The authors formalize this as a learnable capability and train models at 4B, 9B, and 27B scales, comparing a jointly trained omni model…

Cong Wan, Zeyu Guo, Zijian Cai, Jiangyang Li, et al.

Published: Jun 2026
Upvotes: 75
Citations: 0

Read summary Original

59

Research paper

DragMesh-2: Physically Plausible Dexterous Hand-Object Interaction with Articulated Objects

DragMesh-2 is a contact-driven framework for dexterous hand interaction with articulated objects, where the target part moves only through physical hand–handle contact. It extends prior object-centric generation to hand-driven interaction, addressing the challenge that policies trained under fixed dynamics overfit nominal contact loads and degrade under…

Tianshan Zhang, Yijia Duan, Yanjun Li, Zeyu Zhang, et al.

Published: Jun 2026
Upvotes: 74
Citations: 0

Read summary Original

60

Research paper

Latent Spatial Memory for Video World Models

This paper introduces latent spatial memory, a persistent 3D cache for video world models that stores scene information directly in the diffusion latent space, avoiding the pixel-space round trip of RGB point-cloud memories. The authors propose Mirage, a framework that constructs the memory by lifting latent tokens into 3D via depth-guided back-projection…

Weijie Wang, Haoyu Zhao, Yifan Yang, Feng Chen, et al.

Published: Jun 2026
Upvotes: 71
Citations: 2

Read summary Original

61

Research paper

Agentic Environment Engineering for Large Language Models: A Survey of Environment Modeling, Synthesis, Evaluation, and Application

This paper surveys the field of agentic environment engineering for large language models (LLMs), covering the full lifecycle of environment modeling, synthesis, evaluation, and application. It defines agentic environments as dynamic, interactive systems for training and evaluating LLM agents, contrasting them with traditional RL simulators. The survey…

Jiachun Li, Zhuoran Jin, Tianyi Men, Yupu Hao, et al.

Published: Jun 2026
Upvotes: 71
Citations: 1

Read summary Original

62

Research paper

Claw-SWE-Bench: A Benchmark for Evaluating OpenClaw-style Agent Harnesses on Coding Tasks

Claw-SWE-Bench is a multilingual SWE-bench-style benchmark and adapter protocol for evaluating general-purpose agent harnesses (claws) on coding tasks. It comprises 350 GitHub issue-resolution instances across 8 languages and 43 repositories, drawn from SWE-bench-Multilingual and SWE-bench-Verified-Mini after future-commit cleanup. The benchmark fixes…

Mengyu Zheng, Kai Han, Boxun Li, Haiyang Xu, et al.

Published: Jun 2026
Upvotes: 71
Citations: 2

Read summary Original

63

Research paper

Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference

The paper introduces Retrospective Harness Optimization (RHO), a self-supervised method to improve an AI agent's harness (skills, tools, workflows) using only past trajectories, without ground-truth labels. RHO selects a diverse, challenging coreset of past tasks via a determinantal point process, re-solves them in parallel, and extracts improvement…

Wenbo Pan, Shujie Liu, Chin-Yew Lin, Jingying Zeng, et al.

Published: Jun 2026
Upvotes: 70
Citations: 1

Read summary Original

64

Research paper

KVarN: Variance-Normalized KV-Cache Quantization Mitigates Error Accumulation in Reasoning Tasks

The paper introduces KVarN, a calibration-free KV-cache quantization method that mitigates error accumulation during autoregressive decoding in long-horizon reasoning tasks. It identifies that quantization errors accumulate across timesteps, driven primarily by incorrect token scales, which cause outlier errors that disproportionately degrade end-to-end…

Lorenz K. Muller, Philippe Bich, Chiara Boretti, Hyun-Min Chang, et al.

Published: Jun 2026
Upvotes: 69
Citations: 3

Read summary Original

65

Research paper

LatentSkill: From In-Context Textual Skills to In-Weight Latent Skills for LLM Agents

LatentSkill is a framework that converts textual agent skills into plug-and-play LoRA adapters via a pretrained hypernetwork, storing skill knowledge in weight space rather than context space. This removes per-step skill tokens from prompts while preserving modular loading, scaling, and composition. On ALFWorld, LatentSkill improves success by 21.4 and…

Aofan Yu, Chenyu Zhou, Tianyi Xu, Zihan Guo, et al.

Published: Jun 2026
Upvotes: 67
Citations: 2

Read summary Original

66

Research paper

DomainShuttle: Freeform Open Domain Subject-driven Text-to-video Generation

DomainShuttle is a novel framework for open-domain subject-driven text-to-video (S2V) generation, addressing both in-domain (high subject fidelity) and cross-domain (flexible adaptation of subject-irrelevant features) scenarios. It introduces three key components: Domain-MoT, which decouples video and reference features and uses domain-aware AdaLN for…

Nan Chen, Yiyang Cai, Rongchang Xie, Junwen Pan, et al.

Published: Jun 2026
Upvotes: 67
Citations: 1

Read summary Original

67

Research paper

Z-Reward: Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions

Z-Reward is a teacher-student framework for text-to-image reward modeling that represents visual preference as a reasoning-conditioned score distribution rather than a scalar. The teacher, a 27B VLM, is trained with Group-wise Direct Score Optimization (GDSO), which combines GRPO-style policy-gradient rewards with direct supervision on score distributions…

Xin Jin, Huanqia Cai, Zhen Li, Zechao Zhan, et al.

Published: Jun 2026
Upvotes: 66
Citations: 0

Read summary Original

68

Research paper

FlashMemory-DeepSeek-V4: Lightning Index Ultra-Long Context via Lookahead Sparse Attention

This report introduces FlashMemory-DeepSeek-V4 (FM-DS-V4), which uses Lookahead Sparse Attention (LSA) to reduce GPU memory for ultra-long context LLM serving. LSA adds a Neural Memory Indexer to DeepSeek-V4-Flash, predicting and fetching only critical KV chunks every 64 steps, instead of keeping the full KV cache. The indexer is trained independently as a…

Yan Wang, Qifan Zhang, Jiachen Yu, Tian Liang, et al.

Published: Jun 2026
Upvotes: 66
Citations: 2

Read summary Original

69

Research paper

PerceptionDLM: Parallel Region Perception with Multimodal Diffusion Language Models

The paper introduces PerceptionDLM, a multimodal diffusion language model for efficient parallel region perception. It builds on PerceptionDLM-Base, a strong diffusion-based vision-language model, and adds region prompting, RoI-aligned feature replay, and structured attention masking to generate captions for multiple image regions simultaneously in a…

Yueyi Sun, Yuhao Wang, Jason Li, Ye Tian, et al.

Published: Jun 2026
Upvotes: 65
Citations: 0

Read summary Original

70

Research paper

NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?

NatureBench is a benchmark of 90 tasks distilled from peer-reviewed Nature-family publications (2022–2025) to evaluate whether AI coding agents can move beyond reproduction toward scientific discovery. It is built on NatureGym, an automated pipeline that converts papers into containerized task packages with an information firewall, hidden ground truth, and…

Yuru Wang, Lejun Cheng, Yuxin Zuo, Sihang Zeng, et al.

Published: Jun 2026
Upvotes: 65
Citations: 1

Read summary Original

71

NVIDIA

Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients

The paper introduces Zone of Proximal Policy Optimization (ZPPO), a post-training method for small vision-language models (VLMs) that transfers knowledge from a larger teacher without imitating its logits or injecting its responses into the policy gradient. ZPPO addresses two failure modes: distillation's brittleness in the small-student regime and RL's…

Byung-Kwan Lee, Ximing Lu, Shizhe Diao, Minki Kang, et al.

Published: Jun 2026
Upvotes: 64
Citations: 2

Read summary Original

72

Research paper

In-Context World Modeling for Robotic Control

The paper introduces In-Context World Modeling (ICWM), a framework that enables Vision-Language-Action (VLA) models to adapt to novel system configurations (e.g., camera viewpoints, robot morphologies) without parameter updates. Standard VLA models condition only on current observations and language instructions, implicitly assuming a fixed execution…

Siyin Wang, Junhao Shi, Senyu Fei, Zhaoyang Fu, et al.

Published: Jun 2026
Upvotes: 63
Citations: 0

Read summary Original

73

Research paper

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses

Harness-1 is a 20B search agent trained with reinforcement learning inside a stateful search harness. The harness maintains environment-side working memory (candidate pool, importance-tagged curated set, evidence graph, verification records, compressed/deduplicated observations, budget-aware rendering), while the policy retains semantic decisions: what to…

Pengcheng Jiang, Zhiyi Shi, Kelly Hong, Xueqiang Xu, et al.

Published: Jun 2026
Upvotes: 60
Citations: 2

Read summary Original

74

bioRxiv

Measuring Epistemic Resilience of LLMs Under Misleading Medical Context

MedMisBench is a benchmark introduced to measure the epistemic resilience of large language models (LLMs) in medical settings, defined as the ability to maintain correct medical judgment when misleading context is present. It contains 10,932 medical question items and 48,889 misleading context-option pairs, built from five source datasets covering medical…

Hongjian Zhou, Xinyu Zou, Jinge Wu, Sean Wu, et al.

Published: Jun 2026
Upvotes: 60
Citations: 1

Read summary Original

75

Research paper

Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages

Multi-LCB is a benchmark extending LiveCodeBench (LCB) to twelve programming languages (C++, C#, Python, Java, Rust, Go, TypeScript, JavaScript, Ruby, PHP, Kotlin, Scala) to evaluate LLM code generation beyond Python. It converts LCB's Python tasks into a unified STDIN/STDOUT format, preserving contamination controls and evaluation protocol, and…

Maria Ivanova, Pavel Zadorozhny, Rodion Levichev, Ivan Petrov, et al.

Published: Jun 2026
Upvotes: 60
Citations: 1

Read summary Original

76

Research paper

K-BrowseComp: A Web Browsing Agent Benchmark Grounded in Korean Contexts

K-BROWSECOMP is a new benchmark for evaluating web-browsing agents in Korean contexts, consisting of 400 problems. The 300-problem K-BROWSECOMP-VERIFIED subset was manually constructed and validated by native Korean speakers. On this subset, frontier LLMs like GPT-5.5, DeepSeek-V4-Pro, and GLM-5.1 achieve only 30.00–45.67% accuracy, a significant drop from…

Nahyun Lee, Dongkeun Yoon, Guijin Son, Geewook Kim, et al.

Published: Jun 2026
Upvotes: 59
Citations: 1

Read summary Original

77

Research paper

LabVLA: Grounding Vision-Language-Action Models in Scientific Laboratories

LabVLA addresses the gap between AI-driven scientific reasoning and physical laboratory execution by introducing a Vision-Language-Action (VLA) model trained for scientific laboratory protocols. The authors identify data and embodiment as central bottlenecks, addressed via RoboGenesis, a simulation-based data engine that generates LabEmbodied-Data, a…

Baochang Ren, Xinjie Liu, Xi Chen, Yanshuo Liu, et al.

Published: Jun 2026
Upvotes: 59
Citations: 0

Read summary Original

78

Research paper

GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?

GameCraft-Bench is a benchmark for evaluating coding agents on end-to-end game generation in the Godot engine. It formalizes the task as transforming natural-language specifications into complete, playable game projects, requiring three desiderata: Engine Grounding, Artifact Completeness, and Interactive Verification. The benchmark includes 140 tasks…

Tongxu Luo, Rongsheng Wang, Jiaxi Bi, Chenming Xu, et al.

Published: Jun 2026
Upvotes: 59
Citations: 1

Read summary Original

79

Research paper

Where Do Deep-Research Agents Go Wrong? Span-Level Error Localization in Agent Trajectories

This paper introduces TELBENCH, a benchmark for span-level error localization in deep-research agent trajectories, and DRIFT, a claim-centric auditing framework. The authors collected 2,790 real trajectories from two agent frameworks (MiroFlow, OAgent), three backbone models (GPT-5, Gemini-2.5-Pro, Claude-Sonnet-4.5), and three benchmarks (GAIA, XBench,…

Jiaming Wang, Ziteng Feng, Jiangtao Wu, Ruihao Li, et al.

Published: Jun 2026
Upvotes: 58
Citations: 0

Read summary Original

80

Research paper

OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning

OPID (On-Policy Skill Distillation) is a framework for agentic reinforcement learning that extracts hierarchical hindsight skills from completed on-policy trajectories to provide dense token-level supervision, complementing sparse outcome-based RL. It represents trajectory hindsight as episode-level skills (global workflows or failure-avoidance rules) and…

Shuo Yang, Jinyang Wu, Zhengxi Lu, Yuhao Shen, et al.

Published: Jun 2026
Upvotes: 57
Citations: 8

Read summary Original

81

Research paper

SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research

SearchSwarm introduces a method for training delegation intelligence in LLM agents for long-horizon deep research tasks. The approach uses a main-distributes, sub-executes paradigm where a main agent decomposes tasks and dispatches subtasks to subagents via a call_sub_agent tool, receiving only condensed, citation-grounded reports. The harness design…

Pu Ning, Quan Chen, Kun Tao, Xinyu Tang, et al.

Published: Jun 2026
Upvotes: 56
Citations: 3

Read summary Original

82

Research paper

World Action Models: A Survey

This survey defines World Action Models (WAMs) as embodied predictive-action models that make a forecast of the future available to action, distinguishing them from VLAs, world models, and video generation models. It organizes WAMs via two complementary views: a design-philosophy taxonomy (Render-and-Decode, Latent-Only, Video-Generation-Free) and a…

Qiuhong Shen, Shihua Zhang, Yue Liao, Qi Li, et al.

Published: Jun 2026
Upvotes: 56
Citations: 1

Read summary Original

83

Research paper

SoCRATES: Towards Reliable Automated Evaluation of Proactive LLM Mediation across Domains and Socio-cognitive Variations

SoCRATES is a benchmark for evaluating proactive LLM mediators in realistic, multi-domain conflict scenarios. It uses an agentic pipeline to curate scenarios from real conflicts across eight domains, probes mediators along five socio-cognitive axes (strategic posture, party composition, history length, emotional reactivity, cultural identity), and scores…

Taewon Yun, Hyeonseong Park, Jeonghwan Choi, Hayoon Park, et al.

Published: Jun 2026
Upvotes: 55
Citations: 0

Read summary Original

84

Research paper

ACE-Ego-0: Unifying Egocentric Human and Robotic Data for VLA Pretraining

ACE-EGO-0 is a unified Vision-Language-Action (VLA) pretraining framework that jointly trains on heterogeneous embodied data, including egocentric human videos, multi-embodiment robot demonstrations, and simulation rollouts. It addresses representation heterogeneity via a unified action representation: camera-space actions (spatial alignment),…

Hao Li, Ganlong Zhao, Yufei Liu, Haotian Hou, et al.

Published: Jun 2026
Upvotes: 55
Citations: 1

Read summary Original

85

Z.ai / GLM

SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning

SCAIL-2 is an end-to-end framework for controlled character animation that bypasses intermediate representations like pose skeletons or masked backgrounds, which cause information loss. It directly concatenates driving videos to the sequence, allowing the model to capture all visual information. To address the lack of end-to-end data, the authors unify…

Wenhao Yan, Fengjia Guo, Zhuoyi Yang, Jie Tang

Published: Jun 2026
Upvotes: 54
Citations: 1

Read summary Original

86

Together AI

ParallelKernelBench: Benchmarking LLMs on Multi-GPU Kernel Generation

The paper introduces ParallelKernelBench (PKB), a benchmark for evaluating LLMs on multi-GPU CUDA kernel generation. PKB includes 87 problems spanning parallelism strategies (tensor, expert, data, sequence, context) and real workloads from production frameworks. Evaluations of frontier models (GPT-5.5, Opus-4.7, Gemini 3 Pro, GLM-5.2, DeepSeek V4 Pro) show…

Willy Chan, Nathan Paek, Simon Guo, Simran Arora, et al.

Published: Jun 2026
Upvotes: Not indexed
Citations: Not indexed

Read summary Original