New
Everything new in AI as it appears: launches announced by the labs, brand-new models, datasets, papers and open-source tools. Checked every minute or two. Launches feed · Trending · News
Live now
- Connecting…
Thursday 22 October
- DatasetSTT-DATASET/en-accent-arl2eng-arabic-l1
arl2eng_arabic_l1 ASR-ready version of ArL2Eng (figshare 27893778) (license: CC-BY-4.0). Arabic-L1 speakers reading the Speech Accent Archive paragraph. Labels = Soniox t
- DatasetSamChen567/VideoBenchEval-results
phase3_qbase_full_hybrid_lr1e-5：評測與 ablation 結果 截至 2026-10-23（UTC+8）。模型：SamChen567/phase3_qbase_full_hybrid_lr1e-5。評測程式快照在 code/（VideoBenchEval：vbe/infer.py 主評測、vbe/ablat
- Datasetfazmain/parent-taught-refusal-teacher-swap
Teacher-swap test: run outputs Run outputs of the first extension experiment to the paper "Refusal survives embodiment: a parent-taught verdict for a vision-language-acti
- Dataset
- Dataset
- DatasetBeifengdo/Fundamentals-of-Control-Theory-Spring-2019
version https://git-lfs.github.com/spec/v1 oid sha256:5b7cd4ad373f35b76d466b3930ad19b9c01a63a2e641a706a64691d599ea532b size 466
- Datasetkeyur-neeved/DF-Mix-filtered
DF-Mix-filtered Filtered version of keyur-neeved/DF-Mix (commit b5266b2f), same columns and labels: class_id, class_name, binary_label, binary_name, image. class_id class
- DatasetThaumaturgicon7/hedge-density-atlas
Hedge Density Atlas — Scored LLM Output Benchmark Forensic, evidence-based benchmark of production LLM output: epistemic hedging and structural density, measured determin
- Datasetsadaasf/Annoy-PyEdu-Rs-Raw
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- Datasetsadaasf/Annoy-PyEdu-Rs
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- DatasetAI4Manufacturing/208-mcq
Dataset Card Dataset Description [Placeholder] Dataset Structure [Placeholder] Uses [Placeholder] Limitations [Placeholder] License [Placeholder] Citation [Placeholder]
- Dataset
- DatasetASDSAD2E12EDASD/Annoy-PyEdu-Rs-Raw
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- DatasetASDSAD2E12EDASD/Annoy-PyEdu-Rs
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- Datasetzils-ai/base-exam-seed-v1
Zils base-exam public seed v1 Public development sample for the Zils decision-quality base election (zils.base_exam seed bundle, manifest.json). 295 decision items across
- Datasetmaxwildersmith/splataverse-nextgenminimal
Splataverse NextGenMinimal Large Unreal-Engine-derived scenes (USD) converted to tiled Gaussian splats (.spz, SPZ v4, degree-0 SH). scene Gaussians tiles tile edge size n
- DatasetAmeyapores/haply_stack_yellow_red
This dataset was created using LeRobot. Dataset Structure meta/info.json: { "codebase_version": "v3.0", "robot_type": "franka", "total_episodes": 35, "total_frames": 1938
- DatasetAI4Manufacturing/208-region
Dataset Card Dataset Description [Placeholder] Dataset Structure [Placeholder] Uses [Placeholder] Limitations [Placeholder] License [Placeholder] Citation [Placeholder]
- Datasetlichtphyz/rollout_act_v3_aug_desktop
This dataset was created using LeRobot. Dataset Description Part of the SO-101 block-to-bin collection (models, training data and real-robot evaluations). Real-robot eval
- DatasetAI4Manufacturing/208-grounding
Dataset Card Dataset Description [Placeholder] Dataset Structure [Placeholder] Uses [Placeholder] Limitations [Placeholder] License [Placeholder] Citation [Placeholder]
- DatasetAI4Manufacturing/185-grounding
Dataset Card Dataset Description [Placeholder] Dataset Structure [Placeholder] Uses [Placeholder] Limitations [Placeholder] License [Placeholder] Citation [Placeholder]
- Datasetaravindsuresh16/cathlab_handover_mujoco_v1
This dataset was created using LeRobot. Dataset Structure meta/info.json: { "codebase_version": "v3.0", "fps": 30, "features": { "observation.state": { "dtype": "float32"
- DatasetAI4Manufacturing/184-grounding
Dataset Card Dataset Description [Placeholder] Dataset Structure [Placeholder] Uses [Placeholder] Limitations [Placeholder] License [Placeholder] Citation [Placeholder]
- Dataset
- Datasetykeee/anycam-mvhumannet-hdri-orbit-wan22-5b-latents
MVHumanNet HDRI orbit latent bank — Wan2.2 5B Incremental upload in progress. Published: 945 verified outfits, 11340 source/target pairs. The current manifest.json lists
- Dataset
- Datasetskumar-ml/belief-world-models
BB-WM agent trajectories Trajectories for Towards a Belief-Based World Model for LLM Agents (BB-WM). ReAct and ReflAct agents are evaluated on ALFWorld, ScienceWorld, and
- Dataseth2oai/h2o-lightning-card-assets
H2O-Lightning card assets The images shown on the H2O-Lightning model cards, kept here so that they are readable without logging in (the model repositories they belong to
- DatasetBeifengdo/Postgraduate-Politics-Xiao-Xiurong
version https://git-lfs.github.com/spec/v1 oid sha256:e9656c572468de395510383860817e7a57618b09a3f29e53c0904c53e4a6967b size 463
- Datasetbackpr/us-youth-program-data
BackPR open data for US youth programs Reference data on camps, after-school programs, childcare and youth sports in the United States: state rules for all 50 states and
- Datasetburstiness-engine/sentence-length-dispersion
Sentence-Length Dispersion: Human Articles and Model Generations The data behind Can Prompting Reliably Close the Sentence-Length Dispersion Gap? (Vittoria Lanzo and Dico
- Dataset
- Dataset
- DatasetSrushrey/Figshare_Brain_Tumor
Figshare Brain Tumor Dataset (Cheng et al. 2015) Brain tumor MRI dataset from Cheng et al., hosted on Figshare. 3064 T1-weighted contrast-enhanced MRI slices from 233 pat
- Dataset
- Dataset
- Dataset
- DatasetASCXZ12EWDX/Annoy-PyEdu-Rs-Raw
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- DatasetASCXZ12EWDX/Annoy-PyEdu-Rs
Annoy: This should be a paper Title 📑 Paper | 🌐 Project Page | 💾 Released Resources | 📦
- DatasetMuxinZhou/mapladder-town15
MapLadder areas: 8 Town15 areas, their five ladder human maps and the truth Eight 250 m areas of CARLA Town15 as seen in Isaac Sim 5.1, each with its input files, its fiv
Headlines and links come from the labs' own blogs, Hacker News and the GDELT news index, and from the sources our index already follows. We show a short excerpt and always link to the original. A launch reported by a news site rather than the lab is marked “Reported” until the lab itself announces it.