No spam, ever. We'll never share your email address and you can opt out at any time. Already a subscriber? Log in
๐ฃ Announcing the release of the ๐๏ธ Annotated Encyclical ๐๏ธ from the ethics & society folks at @hf.co. Includes citations to relevant academic work. Very much a WIP. Please add work we haven't added yet! huggingface.co/spaces/socie...
๐ฅ Our latest work is state of the art 3D pose estimation, with video analysis from a single camera! ๐ฅ Just train with 3D data, and then only 1 camera is needed! ๐ค Presented this week at #CVPR2026 by lead author Ti Wang! Please repost๐ฉต๐๐ผ xiu-cs.github.io/FMPose3D/ arxiv.org/absโฆ
Coming back from #CVPR2026 in Denver after spending almost a week. I gave a short oral talk this year at the @how_cvpr2026 on "A Case Study on Hidden Bias in Vision-Language Model Activations." Feel free to check out the paper recently published with Mahzarin Banaji!! arxiv.orโฆ
Introducing RoMo, a dataset of 820K high-quality 3D human motion clips, featuring a detailed semantic taxonomy. This innovation bridges the motion data gap, allowing generative models to achieve high fidelity and nuanced understanding of complex prompts. https://arxiv.org/abs/โฆ
A study shows that vision-language models rely on visual encoders for spatial reasoning in tasks like image captioning. By enhancing vision-derived ordering signals, researchers achieved improvements in model performance, paving the way for more efficient AI. https://arxiv.orgโฆ
Research exposes flaws in Vision-Language Models' chronological reasoning, revealing their reliance on superficial cues like grayscale images over accurate historical semantics. New datasets seek to guide the development of more logically driven multimodal AI. https://arxiv.orโฆ
In the text, Leo denounced the โculture of powerโ driving the AI race, especially in developing ever more sophisticated methods of remote warfare. ... "Artificial Intelligence now demands to be disarmed, freed from logics that turn it into an instrument of domination, exclusioโฆ
ByteDance dropped Lance๐ huggingface.co/bytedance-re... This 3B model can generate images + edit images + generate videos + edit videos, and understand both images/videos. It's trained from scratch on only 128 A100s, and beats several 7B+ models on GenEval and VBench!
Meta Lays Off 8,000 Employees, as A.I. Casualties Mount www.nytimes.com/2026/05/19/t...
5/ MVEB: Massive Video Embedding Benchmark โ as AI-generated video proliferates, good video embeddings become critical. You can't grep through video. https://arxiv.org/abs/2606.14958 https://x-cancel.com/Muennighoff/status/2066937509050024254
4/ WEAVER: open-source world model for robot manipulation (weights, code, paper) โ designed for policy evaluation, improvement & test-time planning with limited real-world interaction. https://arxiv.org/abs/2606.13672 https://x-cancel.com/andrea_bajcsy/status/2067255765438538151
๐๐๐ฅ๐: ๐๐ฎ๐๐ฒ๐ป๐ ๐๐ฐ๐๐ถ๐ผ๐ป ๐ฅ๐ฒ๐ฝ๐ฟ๐ฒ๐๐ฒ๐ป๐๐ฎ๐๐ถ๐ผ๐ป ๐๐น๐ถ๐ด๐ป๐บ๐ฒ๐ป๐ ๐ณ๐ผ๐ฟ ๐ฉ๐ถ๐๐ถ๐ผ๐ป-๐๐ฎ๐ป๐ด๐๐ฎ๐ด๐ฒ-๐๐ฐ๐๐ถ๐ผ๐ป ๐ ๐ผ๐ฑ๐ฒ๐น๐ Mengya Liu, Baoxiong Jia, Jiangyong Huang ... Siyuan Huang arxiv.org/abs/2606.07100 Trending on scholar-inbox.com
I watch the discussion over Anthropic on Bluesky with ๐ฟ๐ฟ๐ฟ
Suddenly everyone's algorithms started "self-improving"... Like, there are no human researchers working in the companies, or thousands of raters for RLHF. LIARS
Center = Evangelos Kazakos. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.