[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@HuggingPapers "Mistral released Voxtral on Hugging Face Incorporates state-of-the-art audio input capabilities into LLMs while retaining best-in-class text performance. 3B and 24B variants It excels at speech transcription translation and audio understanding"
@HuggingPapers on X 2025-07-16 15:50:44 UTC 3764 followers, XXX engagements

"Ant Group researchers just dropped GUI-G on Hugging Face This new framework for GUI grounding uses Gaussian reward modeling transforming sparse binary classification to dense continuous optimization. Achieves state-of-the-art results with a XXXX% boost on ScreenSpot-Pro"
@HuggingPapers on X 2025-07-22 12:10:35 UTC 3757 followers, XXX engagements

"Qualcomm AI Research just dropped CSD-VAR A novel method for content-style decomposition in visual autoregressive models. It leverages VAR's scale-wise generation for superior disentanglement and creative flexibility"
@HuggingPapers on X 2025-07-22 00:21:33 UTC 3776 followers, 8854 engagements

"Qwen just released their updated Qwen3-235B-Instruct on Hugging Face. It boasts significant gains in instruction following logical reasoning and 256K long-context understanding"
@HuggingPapers on X 2025-07-21 17:38:37 UTC 3778 followers, 8741 engagements

"Mistral AI just dropped Voxtral A X B & XX B open-source speech-understanding model that beats Whisper large-v3 GPT-4o mini and Gemini XXX Flash on transcription and understanding while costing of comparable APIs"
@HuggingPapers on X 2025-07-20 16:06:05 UTC 3773 followers, 1601 engagements

"Google Research and collaborators unveil AgentsNet A new benchmark for multi-agent LLM coordination Tests how LLMs collaborate in networks scaling up to XXX agents"
@HuggingPapers on X 2025-07-17 00:20:20 UTC 3765 followers, 9661 engagements

"Qwen just dropped Qwen3-Coder on Hugging Face It's their most agentic code model yet with 256K native context (1M with Yarn) and performance comparable to Claude Sonnet on agentic coding tasks"
@HuggingPapers on X 2025-07-22 21:36:13 UTC 3772 followers, 3035 engagements

"MMHU A massive-scale multimodal benchmark for human behavior understanding in autonomous driving. 57k human motion clips & 1.73M frames from Waymo YouTube & self-collected data with rich annotations for motion trajectory intention & safety-critical labels"
@HuggingPapers on X 2025-07-17 18:41:55 UTC 3772 followers, 8326 engagements

"ByteDance's MiroMind AI just released MiroMind-M1 This open-source RLM series built on Qwen-2.5 achieves state-of-the-art mathematical reasoning. It uses a novel Context-Aware Multi-Stage Policy Optimization"
@HuggingPapers on X 2025-07-22 08:09:28 UTC 3760 followers, 1283 engagements

"New paper: MegaScience is pushing the frontiers of science reasoning They introduce new datasets (TextbookReasoning MegaScience) totaling 1.25M instances. Models trained on them significantly outperform official instruct counterparts"
@HuggingPapers on X 2025-07-23 12:10:33 UTC 3772 followers, XXX engagements

"New paper just dropped on Hugging Face: a statistical diagnosis for training LLM web agents. It shows how combining SFT with on-policy RL achieves SOTA performance using only XX% of the compute required by pure SFT"
@HuggingPapers on X 2025-07-12 08:07:30 UTC 3775 followers, XXX engagements

"MoVieS reconstructs a full 4D dynamic scene from a single monocular video in just one second No optimization loops no test-time trainingjust feed-forward speed with pixel-wise motion geometry and appearance in one shot"
@HuggingPapers on X 2025-07-20 04:21:28 UTC 3772 followers, XXX engagements

"Dive deeper into its capabilities for repository-scale understanding and impressive agentic coding. Check out the model and paper for more details Model: Paper:"
@HuggingPapers on X 2025-07-22 21:36:15 UTC 3772 followers, XXX engagements

"Learn how to train your LLM web agents more efficiently and close the gap with closed-source models. Read the full paper on Hugging Face:"
@HuggingPapers on X 2025-07-12 08:07:32 UTC 3760 followers, XXX engagements

"Qwen just released the Qwen3 235B Instruct model on Hugging Face This updated FP8 MoE model brings significant improvements in reasoning coding and multi-language capabilities alongside enhanced 256K long-context understanding"
@HuggingPapers on X 2025-07-22 07:05:33 UTC 3761 followers, XXX engagements

"NVIDIA just dropped Llama-Nemotron Super 49B on Hugging Face This new model a derivative of Llama-3.3 excels in reasoning chat and agentic tasks with a massive 128K context window. It offers a powerful balance of accuracy and efficiency"
@HuggingPapers on X 2025-07-26 00:10:33 UTC 3778 followers, XXX engagements

"Designing high-quality slides Sony & KAIST introduce DesignLab: an AI framework that iteratively detects & corrects design issues turning your drafts into polished presentations"
@HuggingPapers on X 2025-07-24 08:12:28 UTC 3778 followers, XXX engagements

"Open Vision Reasoner (OVR) is here. A new framework for multimodal LLMs that transfers linguistic cognitive behaviors to unlock advanced visual reasoning achieving SOTA performance"
@HuggingPapers on X 2025-07-14 20:07:34 UTC 3765 followers, XXX engagements

"Alibaba Group and UIUC introduce PAPO a new RL framework for multimodal reasoning. It teaches models to perceive & reason by tackling the visual perception bottleneck. XXXX% reduction in perception errors. XXX% overall gain. No extra data needed"
@HuggingPapers on X 2025-07-10 04:13:55 UTC 3773 followers, 1030 engagements

"New paper introduces RALU: Region-Adaptive Latent Upsampling This training-free method dramatically accelerates Diffusion Transformers delivering up to 7x speedup on FLUX & 3x on Stable Diffusion X. It upsamples only what matters retaining stunning image quality"
@HuggingPapers on X 2025-07-23 16:10:36 UTC 3777 followers, 14.6K engagements

"AllenAI just soft-launched a new olmOCR model on Hugging Face Be among the first to explore their latest advancement in OCR technology"
@HuggingPapers on X 2025-07-23 15:38:59 UTC 3773 followers, XXX engagements

"Discover the challenges of multi-agent LLM coordination as networks scale. Explore interactive demos and dive into the data Paper: Dataset: Demo:"
@HuggingPapers on X 2025-07-17 00:20:22 UTC 3759 followers, XXX engagements

"ByteDance just released VINCIE on Hugging Face. This groundbreaking model unlocks in-context image editing directly from video setting new SOTA on multi-turn benchmarks. Transform your images with ease"
@HuggingPapers on X 2025-07-02 18:23:49 UTC 3769 followers, 21.1K engagements

"A 165-page survey distilling 1300+ papers into Context Engineering - the discipline that moves AI from static prompts to dynamic production-grade context orchestration"
@HuggingPapers on X 2025-07-18 07:13:41 UTC 3774 followers, 1497 engagements

"ByteDance Seed team just unveiled GR-3: a new large-scale Vision-Language-Action (VLA) model for generalist robots. It shows exceptional generalization to novel objects and instructions handling complex long-horizon tasks"
@HuggingPapers on X 2025-07-23 00:20:59 UTC 3772 followers, XXX engagements

"NVIDIA just released Long-RL It's a full-stack framework Scaling reinforcement learning to long videos up to 256k tokens on a single A100 node"
@HuggingPapers on X 2025-07-11 04:19:46 UTC 3760 followers, 8586 engagements

"GR-3 is robust reliable and efficiently fine-tuned with minimal human data. It outperforms SOTA baselines in real-world experiments. Dive into the technical report: See GR-3 in action:"
@HuggingPapers on X 2025-07-23 00:21:01 UTC 3772 followers, XXX engagements