Explore · 每周精选
发现最受关注的研究论文,追踪研究趋势,订阅感兴趣的期刊与关键词。
Maijunxian Wang, Ruisi Wang, Juyi Lin +53 authors
A large-scale video reasoning dataset and benchmark are introduced to study video intelligence capabilities beyond visual quality, enabling systematic analysis of spatiotemporal reasoning and generalization across diverse tasks.
50 篇论文 · 按点赞排序
Shaobo Wang, Xuan Ouyang, Tianyi Xu +9 authors
OPUS is a dynamic data selection framework that improves pre-training efficiency by scoring data candidates based on optimizer-induced update projections in a stable proxy-derived target space, achieving superior performance with reduced computational overhead.
I. Apanasevich, M. Artemyev, R. Babakyan +25 authors
Green-VLA is a five-stage vision-language-action framework for real-world robot deployment that achieves generalization across different robot embodiments through multimodal training and reinforcement learning.
Zehao Chen, Gongxun Li, Tianxiang Ai +8 authors
WMSS is a post-training paradigm that uses weak model checkpoints to identify and fill learning gaps, enabling continued improvement beyond conventional saturation points in large language models.
Kimi Team, Tongtong Bai, Yifan Bai +323 authors
Kimi K2.5 is an open-source multimodal agentic model that enhances text and vision processing through joint optimization techniques and introduces Agent Swarm for parallel task execution.
Haifeng Wang, Hua Wu, Tian Wu +435 authors
ERNIE 5.0 is a production-scale trillion-parameter autoregressive model that unifies multimodal understanding and generation through sparse MoE architecture and elastic training.
Zixuan Huang, Xin Xia, Yuxi Ren +11 authors
Large reasoning models can implicitly determine optimal stopping points for thinking, which SAGE-RL enhances by incorporating efficient reasoning patterns into pass@1 inference for improved accuracy and efficiency.
Yuejie Li, Ke Yang, Yueying Hua +4 authors
Spoken query retrieval is an important interaction mode in modern information retrieval. However, existing evaluation datasets are often limited to simple queries under constrained noise conditions, making them inadequate for assessing the robustness of spoken query retrieval systems under complex acoustic perturbations. To address this limitation, we present SQuTR, a robustness benchmark for spoken query retrieval that includes a large-scale dataset and a unified evaluation protocol. SQuTR aggregates 37,317 unique queries from six commonly used English and Chinese text retrieval datasets, spanning multiple domains and diverse query types. We synthesize speech using voice profiles from 200 real speakers and mix 17 categories of real-world environmental noise under controlled SNR levels, enabling reproducible robustness evaluation from quiet to highly noisy conditions. Under the unified protocol, we conduct large-scale evaluations on representative cascaded and end-to-end retrieval systems. Experimental results show that retrieval performance decreases as noise increases, with substantially different drops across systems. Even large-scale retrieval models struggle under extreme noise, indicating that robustness remains a critical bottleneck. Overall, SQuTR provides a reproducible testbed for benchmarking and diagnostic analysis, and facilitates future research on robustness in spoken query to text retrieval.
Dawei Zhu, Rui Meng, Yale Song +4 authors
_paperbanana is an agentic framework that automates the creation of publication-ready academic illustrations using advanced vision-language models and image generation techniques.
Guobin Shen, Chenxiao Zhao, Xiang Cheng +2 authors
VESPO addresses training instability in LLM reinforcement learning by using variational formulation with variance reduction to correct policy divergence without length normalization.
GLM-5 Team, Aohan Zeng, Xin Lv +183 authors
GLM-5 advances foundation models with DSA for cost reduction, asynchronous reinforcement learning for improved alignment, and enhanced coding capabilities for real-world software engineering.
Kaijie Zhu, Yuzhou Nie, Yijiang Li +10 authors
TermiGen introduces a pipeline for generating verifiable terminal environments and resilient trajectories to improve open-weight LLMs' ability to execute complex tasks and recover from runtime errors.
Jingxuan Wei, Siyuan Li, Yuhang Xu +21 authors
World Models require three consistency principles—modal, spatial, and temporal—for general artificial intelligence, with a proposed benchmark evaluating multimodal learning systems.
Ailin Huang, Ang Li, Aobo Kong +212 authors
Step 3.5 Flash is a sparse Mixture-of-Experts model that achieves frontier-level agentic intelligence through efficient parameter utilization and optimized attention mechanisms, demonstrating strong performance across multiple benchmarks.
Yuhao Zheng, Li'an Zhong, Yi Wang +6 authors
Code2World enables autonomous GUI agents to predict next visual states through renderable code generation, achieving high visual fidelity and structural controllability while improving navigation performance.
Chenxu Wang, Chaozhuo Li, Songyang Liu +10 authors
Multi-agent LLM systems face fundamental limitations in achieving continuous self-improvement while maintaining safety alignment due to inherent statistical blind spots in isolated evolution.
Jun Han, Shuo Zhang, Wei Li +21 authors
Financial markets are noisy and non-stationary, making alpha mining highly sensitive to noise in backtesting results and sudden market regime shifts. While recent agentic frameworks improve alpha mining automation, they often lack controllable multi-round search and reliable reuse of validated experience. To address these challenges, we propose QuantaAlpha, an evolutionary alpha mining framework that treats each end-to-end mining run as a trajectory and improves factors through trajectory-level mutation and crossover operations. QuantaAlpha localizes suboptimal steps in each trajectory for targeted revision and recombines complementary high-reward segments to reuse effective patterns, enabling structured exploration and refinement across mining iterations. During factor generation, QuantaAlpha enforces semantic consistency across the hypothesis, factor expression, and executable code, while constraining the complexity and redundancy of the generated factor to mitigate crowding. Extensive experiments on the China Securities Index 300 (CSI 300) demonstrate consistent gains over strong baseline models and prior agentic systems. When utilizing GPT-5.2, QuantaAlpha achieves an Information Coefficient (IC) of 0.1501, with an Annualized Rate of Return (ARR) of 27.75% and a Maximum Drawdown (MDD) of 7.98%. Moreover, factors mined on CSI 300 transfer effectively to the China Securities Index 500 (CSI 500) and the Standard & Poor's 500 Index (S&P 500), delivering 160% and 137% cumulative excess return over four years, respectively, which indicates strong robustness of QuantaAlpha under market distribution shifts.
SII-OpenMOSS Team, Donghua Yu, Mingshu Chen +37 authors
MOVA is an open-source model that generates synchronized audio-visual content using a Mixture-of-Experts architecture with 32 billion parameters, supporting image-text to video-audio generation tasks.
Veuns-Team, Changlong Gao, Zhangxuan Gu +24 authors
UI-Venus-1.5 is a unified GUI agent with improved performance through mid-training stages, online reinforcement learning, and model merging techniques.
Wenxuan Huang, Yu Zeng, Qiuchen Wang +12 authors
Vision-DeepResearch introduces a multimodal deep-research paradigm enabling multi-turn, multi-entity, and multi-scale visual and textual search with deep-research capabilities integrated through cold-start supervision and reinforcement learning.
Yifei Wang, Yueqi Wang, Zhenrui Yue +6 authors
FASA is a novel framework that uses query-aware token eviction and functional sparsity in RoPE to reduce KV cache memory usage while maintaining high performance in long-context LLM tasks.
Hongrui Jia, Chaoya Jiang, Shikun Zhang +1 authors
Diagnostic-driven Progressive Evolution enables continuous improvement of large multimodal models through iterative diagnosis and targeted data generation guided by identified weaknesses.
Xiaomin Yu, Yi Xin, Wenjie Zhang +12 authors
Researchers address the modality gap in multimodal learning by proposing a fixed-frame theory and a training-free alignment method that enables efficient scaling of multimodal models using unpaired data.
Issar Tzachor, Dvir Samuel, Rami Ben-Ari
Generative multimodal large language models are adapted for video-text embedding and retrieval through intermediate-layer analysis and text-based alignment without visual supervision.
Yu Zeng, Wenxuan Huang, Zhen Fang +13 authors
Vision-DeepResearch benchmark addresses limitations in evaluating visual-textual search capabilities of multimodal models by introducing realistic evaluation conditions and improving visual retrieval through multi-round cropped-search workflow.
Ximing Lu, David Acuna, Jaehun Jung +12 authors
Golden Goose synthesizes unlimited RLVR tasks from unverifiable internet text by creating multiple-choice question-answering versions of fill-in-the-middle tasks, enabling large-scale training and achieving state-of-the-art results in cybersecurity and other domains.
Zhiheng Song, Jingshuai Zhang, Chuan Qin +6 authors
MobileBench is a scalable benchmark for evaluating LLM-based route-planning agents in real-world scenarios, featuring anonymized user queries and a deterministic sandbox for reproducible testing.
Renjie Pi, Grace Lam, Mohammad Shoeybi +3 authors
Researchers developed a synthetic task generation pipeline and analyzed data strategies to improve terminal agent performance, creating a large-scale dataset and models that outperform larger counterparts on benchmark tests.
Zelai Xu, Zhexuan Xu, Ruize Zhang +7 authors
Multi-agent systems using reinforcement learning enable parallel information seeking with scalable orchestration, achieving performance comparable to larger single agents.
Yuling Shi, Chaoxiang Xie, Zhensu Sun +7 authors
Multimodal large language models can effectively understand source code when represented as compressed images, achieving significant token reduction while maintaining or improving performance on code comprehension tasks.
北京市昌平区探索星信息技术及软件开发工作室
京ICP备2026059466号