Notable data and AI papers

Papers627Labs45

627 papers · 45 labs

Group Sequence Policy Optimization research paper by Alibaba (Qwen), 2025

Alibaba (Qwen) · Inference and efficiency · Jul 24, 2025 · 814 citations · 323 upvotes · unverified

Read on arXiv

What it shows

Group Sequence Policy Optimization (GSPO) is a reinforcement learning algorithm that improves training efficiency and performance of large language models by using sequence-level importance ratios and operations.

By Chujie Zheng, Shixuan Liu, Mingze Li and 9 more · arXiv 2507.18071 · PDF

UnverifiedHugging Face's summary; not yet checked by hand.

More from Alibaba (Qwen)

All 78
Agents and evaluationUnverifiedOct 26, 2026Co-Design Bench: Multimodal Collaborative Coding Evaluation for Agentic ModelsCo-Design Bench evaluates how effectively coding agents update software applications based on user requests with text, images, and other media.Alibaba (Qwen)23 upvotesRead →Multimodal and roboticsUnverifiedOct 24, 2026Qwen-Image-PE: Learning Visual Design for Text-to-Image Prompt EnhancementQwen-Image-PE improves text-to-image prompts via two-stage training with caption-based reverse distillation and groupwise design rewards, scaling across data, model size, and information density.Alibaba (Qwen)23 upvotesRead →Multimodal and roboticsUnverifiedOct 23, 2026World-in-Context: Interactive Video World Modeling with In-Context ConditioningWorld-in-Context unifies view, action, and camera controls via explicit tokens and interaction-aware positional encoding to improve interactive video generation and navigation.Alibaba (Qwen)22 upvotesRead →Multimodal and roboticsUnverifiedOct 22, 2026Beyond Watching and Searching: Benchmarking Agents for Video-centered Deep Research ReportsThis benchmark evaluates how well AI agents can research and synthesize information from videos to produce interactive reports.Alibaba (Qwen)19 upvotesRead →Agents and evaluationUnverifiedOct 22, 2026UserArena: Benchmarking Interactive User Simulation for Long-Horizon Shopping AgentsUserArena introduces a real-world interactive benchmark with long-horizon shopping trajectories and an evaluation framework, alongside On-Policy Trajectory Distillation to improve simulator fidelity for agentic recommendations.Alibaba (Qwen)55 upvotesRead →Inference and efficiencyUnverifiedOct 19, 2026SWIFT: Self-guided Weak-to-strong dIFfusion models via Target extrapolationThis training method uses a smaller, frozen model to guide the training of a larger diffusion model, improving quality without increasing inference costs.Alibaba (Qwen)30 upvotesRead →Multimodal and roboticsUnverifiedOct 14, 2026BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World ReferencesThis benchmark evaluates how well AI models can perform complex 3D creation and editing tasks within the Blender software.Alibaba (Qwen)23 upvotesRead →Applied AIUnverifiedOct 14, 2026CommerceGen-Studio: An End-to-End Framework for Commercial-Grade Main-Image GenerationThis paper presents an end-to-end framework for generating commercial-grade e-commerce product images by combining metadata extraction, layout planning, and multi-condition generation.Alibaba (Qwen)25 upvotesRead →

Other inference and efficiency papers

Topic
Inference and efficiencySep 12, 2023Efficient Memory Management for Large Language Model Serving with PagedAttentionvLLM manages the KV cache like pages of virtual memory, serving models with 2 to 4 times the throughput.UC Berkeley9,965 citations72 upvotesRead →Inference and efficiencyMay 27, 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-AwarenessFlashAttention computes exact attention with far fewer GPU memory reads and writes, making long sequences faster.Stanford University5,830 citations16 upvotesRead →Inference and efficiencyUnverifiedFeb 16, 2025Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse AttentionNSA, a trainable sparse attention mechanism, enhances long-context modeling efficiency without sacrificing performance, achieving improvements in speed and accuracy over full attention models.DeepSeek596 citations157 upvotesRead →Inference and efficiencyUnverifiedApr 7, 2025SmolVLM: Redefining small and efficient multimodal modelsSmolVLM, a series of compact multimodal models, achieves high performance with minimal GPU memory usage, making efficient deployment on mobile and edge devices possible.Hugging Face330 citations212 upvotesRead →Inference and efficiencyUnverifiedOct 17, 2025Chronos-2: From Univariate to Universal ForecastingChronos-2, a pretrained model with a group attention mechanism, achieves state-of-the-art performance in zero-shot univariate, multivariate, and covariate-informed forecasting tasks.Amazon276 citations28 upvotesRead →Inference and efficiencyUnverifiedApr 3, 2025Inference-Time Scaling for Generalist Reward ModelingSelf-Principled Critique Tuning enhances pointwise generative reward modeling for large language models, improving scalability and quality compared to existing methods.DeepSeek259 citations59 upvotesRead →
About this paper
Venue
arXiv.org
Citations
814, 80 influential · Semantic Scholar
Upvotes
323 · Hugging Face
Lab
Alibaba (Qwen) · on Companies · on Quarterly

Changes

CitationsNov 9, 2026Influential citations rose from 78 to 80.
CitationsNov 9, 2026Citations rose from 802 to 814.
CitationsNov 2, 2026Influential citations rose from 69 to 78.
CitationsNov 2, 2026Citations rose from 741 to 802.
CitationsOct 12, 2026Citations rose from 722 to 741.
CitationsOct 5, 2026Influential citations rose from 64 to 69.
CitationsOct 5, 2026Citations rose from 694 to 722.
CitationsSep 28, 2026Citations rose from 688 to 694.
CitationsSep 25, 2026A first count of 64 influential citations.
CitationsSep 25, 2026A first count of 688 citations.
New paperSep 25, 2026Found by the weekly Hugging Face scan; not yet checked by hand.