Skip to content
K = 0.731
All news · Today3 items

#Data and training

All collected and published items, ordered by time. Picks are a selected subset.
8/25Tue
  1. Hugging Face Blog60

    Multiverse Computing proposes Quantization-Aware Healing: 4-bit compressed model beats its bfloat16 original on seven of nine benchmarks

    Multiverse Computing has published a paper proposing Quantization-Aware Healing (QAH). After compressing GPT-OSS 120B to 60B parameters and quantizing it to MXFP4, the method distills directly from the original model before compression rather than from a recovered bfloat16 checkpoint.

    Why it matters: The paper presents a repair recipe that distills from the original model after compression and quantization, with stability comparisons against QAT.

  2. Mistral AI36

    Mistral and HUMAIN partner to advance sovereign AI in Saudi Arabia and the Middle East

    Mistral and HUMAIN have announced a strategic partnership covering AI infrastructure, advanced model development, and AI deployment, initially focused on cybersecurity and voice, with plans for frontier models that perform strongly in Arabic. The partnership is worth hundreds of millions of euros. Mistral will explore using HUMAIN's data-center infrastructure, and the two companies will develop a joint go-to-market strategy for regulated industries in Saudi Arabia.

8/24Mon
  1. Import AI28

    Import AI 470: machines should not have rights; SPADE automates environment generation; Hawkeye builds better GPU kernels

    METR research finds that AI's acceleration of science is uneven: cybersecurity-vulnerability reports sped up considerably from 2025 to 2026, mathematical research accelerated only slightly, and AI research itself showed no measurable acceleration. A university collaboration proposed SPADE, a self-play framework in which LLMs alternately generate executable training environments and solve them. Qwen3-30B-A3B scored 58.3 across the game-environment suite, an 8.1-point gain over its base model.

8/22Sat
  1. Google Research43

    Google introduces Biomarker Discovery Framework to identify candidate biomarkers from wearable-sensor data

    Google has released Biomarker Discovery Framework, a multi-agent system that organizes candidate biomarker prioritization into an iterative, human-supervised research loop combining hypothesis generation, parallel statistical analysis, model training, adversarial validation, and literature reasoning. Across 9,279 participant observations in three cohorts, it identified 41 candidate digital biomarkers for mental health and 25 for metabolic health. One example: variability in sleep duration correlated with PHQ-8 severity (ρ = 0.252).

8/21Fri
  3. Microsoft Research38

    Microsoft releases Skala 1.1 with 2.5× more training data; now integrated into CP2K

    Microsoft Research has released Skala 1.1, a deep-learning DFT functional trained on 2.5 times as much data as the previous public version. It ranks first in 32 of the 55 GMTKN55 benchmark categories, with a weighted mean error of 2.8 kcal/mol. The release is available in CP2K and is being integrated into Psi4, FHI-aims, ORCA, and VASP. Microsoft has also introduced a living benchmark that tracks computational performance across versions.

8/20Thu
  1. Jim Fan · X54

    Jim Fan analyzes repetition, recovery, and UMI in GEN-1.5

    Jim Fan attributes GEN-1.5’s promise to naturally repetitive human demonstrations, retained failures that teach recovery, and UMI, which lets humans collect data directly while wearing a robot gripper. He argues that teleoperation loses physical intuition and that sufficient data can enable zero-shot behaviors such as picking up novel objects without finetuning. He remains cautiously optimistic, noting that in-context learning depends on the distance between training and test conditions and that current demos are too simple to draw conclusions.

8/19Wed
  1. Hugging Face Blog62

    IBM research: how much memory does your agent actually need?

    The IBM research team has published follow-up work on ALTK-Evolve, evaluating eight models on 585 multi-step tasks in AppWorld. It finds that agent memory is not an on/off switch but a dosage that needs to be calibrated for each model.

    Why it matters: A comparison across eight models shows that the right amount of agent memory varies with model capability and offers two transferable ways to control costs: retrieval and prompt caching.

8/18Tue
  1. Hugging Face Blog48

    Hugging Face introduces constraint-aware GPU allocator, improving cluster utilization by up to 33 percentage points

    Hugging Face built a constraint-aware GPU allocator and compared it with a FIFO scheduler across seven benchmark scenarios. With the same hardware and workloads, GPU utilization improved by up to 33 percentage points, while priority-weighted throughput rose in every scenario by up to 105%.

8/17Mon
8/14Fri
  1. Hugging Face Blog62

    Hugging Face releases 2026 summer report on the open-model ecosystem

    Hugging Face has released a report on the open-model ecosystem from January through August 2026, drawing six findings from Hub data. Public model repositories grew from 2.43 million to 2.96 million and datasets from 711,000 to one million, but 85.6% of models have fewer than 200 lifetime downloads, while 1.5% of repositories account for 99.2% of downloads.

    Why it matters: Hugging Face analyzes its Hub data on the open-model ecosystem; findings on downloads versus likes and the scale of Qwen derivatives can inform model selection.

8/13Thu
  1. Hugging Face Blog34

    Hugging Face OlmoEarth Studio adds custom embedding export

    Hugging Face's OlmoEarth Studio can now compute and export embeddings; its code and model weights are public. Through the UI or API, users can select a region, time range, encoder variant (Nano: 128 dimensions and 1.4M parameters; Tiny: 192 dimensions and 6.2M parameters; Base: 768 dimensions and 89M parameters), resolution, and imagery source, then export an int8 Cloud-Optimized GeoTIFF.

8/12Wed
  1. Google Research60

    Google Research proposes knowledge profiling: retrieval, not encoding, is the bottleneck for frontier LLM factuality

    Google Research has proposed a knowledge-profiling framework that distinguishes factual errors caused by encoding failures from those caused by retrieval failures, and released WikiProfile, a benchmark containing 2,150 Wikipedia facts.

    Why it matters: By separating encoding failures from retrieval failures, knowledge profiling offers a new way to diagnose the sources of factual errors in frontier models.

  2. Microsoft Research36

    Microsoft Research introduces CARE-X, a clinically oriented radiology VLM combining auxiliary supervision, reward alignment, and tool-assisted measurement

    Microsoft Research has released CARE-X, a research model and unified chest X-ray vision-language model that can generate reports, make structured diagnostic predictions, and localize lesions. It uses reinforcement learning (DAPO) to reward clinical correctness across multiple tasks.

8/11Tue
8/10Mon
  1. Hugging Face Blog48

    Hugging Face proposes efficient knowledge distillation with offline top-K logits and fused, chunked KL loss

    A new Hugging Face paper proposes two systems improvements that sharply reduce the training cost of distilling LLMs: cache the teacher model's top-100 logits offline so teacher and student do not need to reside in GPU memory simultaneously, and use a fused, chunked KL loss to avoid generating a full vocabulary-by-sequence-length matrix.

8/6Thu
  1. Google DeepMind76

    Google DeepMind open-sources WeatherNext cyclone forecasting model

    Google DeepMind has released WeatherNext, an AI model that achieves state-of-the-art forecasts of cyclone tracks, intensity, and wind-field structure, providing about one additional day of lead time on average—roughly a decade of progress in meteorology.

    Why it matters: The source reports measured gains of about one additional day of warning for cyclone tracks and intensity and opens the model weights and code.

7/31Fri
  1. Google Research62

    Google Research proposes Science One Framework for verifiable autonomous research with Chain-of-Evidence

    Google Research has proposed Chain-of-Evidence (CoE), a framework for verifiability implemented in the autonomous-research prototype Science One Framework and the automated auditing tool CoE Audit.

    Why it matters: The article reports a verifiability framework and audit metrics, helping readers compare autonomous research systems on citation and code consistency.

  2. Microsoft Research71

    Microsoft Research releases Echoverse, an evolving training environment for computer-use agents

    Microsoft Research has introduced Echoverse, a set of 12 training worlds for computer-use agents: 10 full-domain environments and two capability environments. Each consists of an application, tasks, and a database-based verifier.

    Why it matters: Microsoft Research has released 12 computer-use-agent training environments with verifiers and reports comparisons of transfer performance between deep and shallow environments.

7/30Thu
  1. Hugging Face Blog22

    GPU management: why idle GPUs are becoming the new grounded airplanes

    A Hugging Face blog argues that GPU utilization, not model intelligence, is AI's next real constraint. GPU costs accrue by calendar hours, but output is produced only during compute hours; after buying GPUs, companies face the challenge of keeping their clusters busy. Using aviation as an analogy, the article says even fully loaded clusters may waste much of their potential because GPUs run all day while demand does not.

7/26Sun
  1. Berkeley AI Research36

    Berkeley AI Research proposes ABBEL, using belief-state supervision to make LLMs efficient at long-horizon interaction

    Berkeley AI Research has proposed ABBEL, a framework that separates an LLM's recursive summaries into natural-language “belief states” and scores their content with supervised signals to reduce performance loss from context compression in long interactions. On the collaborative-programming benchmark CollabBench, reconstruction-based belief scoring cut the performance gap with a full-context model by about 50%, reduced training from 100 steps to 50, and kept peak context length well below the full-context setup.

7/22Wed
7/21Tue
  1. Hugging Face Blog62

    Hugging Face releases Grabette, an open system for recording robot manipulation data

    Hugging Face has released Grabette, a low-cost, open-source handheld gripper system for recording robot-manipulation demonstrations without a robot or teleoperation rig. It has a wide-angle fisheye camera and an RGB-D camera for policy-view recording and 6-DoF trajectory tracking, respectively. The hardware bill of materials is about €490; the companion robot-side gripper, Gripette, costs about €120.

    Why it matters: The source describes the full open pipeline from handheld-gripper recording to LeRobot datasets, helping readers assess how the barrier to collecting manipulation data is changing.

7/16Thu
7/9Thu
  1. Google Research62

    Google releases SensorFM, a large sensor foundation model for wearable-health data

    Google Research and Google DeepMind have introduced SensorFM, a large sensor foundation model that learns directly from unlabeled wearable data. Its pretraining corpus contains more than one trillion minutes of multimodal sensor signals from five million consenting participants across more than 100 countries, all 50 U.S. states, and over 20 Fitbit and Pixel Watch devices.

    Why it matters: The article reports pretraining-data scale, scaling experiments, and downstream-task comparisons, helping readers assess how generalizable wearable foundation models are.

7/7Tue
  1. Berkeley AI Research40

    What happens when intelligence becomes cheap? Data systems for agents, operated by agents, and built by agents

    Berkeley's EPIC Data Lab argues that as inference costs have fallen from around $30 per million tokens for GPT-4-level models in early 2023 to under $1 today—and below $0.10 at some vendors—data systems face three new challenges: systems designed for agents, systems operated by agents, and systems built by agents.

  2. Hugging Face Blog74

    LeRobot v0.6.0 released: imagine, evaluate, improve

    Hugging Face has released LeRobot v0.6.0, adding world-model policies (VLA-JEPA, FastWAM, LingBot-VA), a new set of VLA models (GR00T N1.7, MolmoAct2, EO-1, EVO1, Multitask DiT), and a unified reward-model API (Robometer, TOPReward) to the robotics-learning loop.

    Why it matters: The release brings world-model policies, a reward-model API, and six simulation benchmarks into one CLI, giving readers a view of how production-ready the robotics-learning loop is becoming.

7/6Mon
  1. Hugging Face Blog38

    PRX Part 4: our data strategy

    Part four of the PRX series describes its data strategy: pretraining data combines public and internal datasets; a VLM regenerates long descriptions of images, which are then converted into a streamable training corpus. The team switched to Qwen3-VL for on-the-fly text-latent computation during training, losing only about 3–4% throughput at 7B scale (adding roughly one day to a 30-day run). All images are encoded as JPEG at quality 92.

7/1Wed
6/30Tue
  1. Google Research76

    Google releases TabFM, a zero-shot foundation model for tabular data

    Google has released TabFM, a zero-shot foundation model for classification and regression on tabular data. It recasts table prediction as an in-context-learning problem, producing predictions in a single forward pass without manual training, tuning, or feature engineering.

    Why it matters: The source details TabFM's architecture and TabArena results, helping readers assess whether zero-shot table prediction can replace traditional tree-model workflows.

6/26Fri
  1. Lilian Weng · X57

    Lilian Weng explains scaling laws and compute-optimal allocation

    Lilian Weng shares a long-overdue (3+ years?) post on scaling laws. Compute is expensive. Scaling laws help reason about optimal compute allocation between data and model size before committing to a large run. The post covers what scaling laws predict, how compute-optimal allocation works, why Kaplan et al. and Chinchilla disagree, and how data limits and fitting details make extrapolation tricky. https://lilianweng.github.io/posts/2026-06-24-scaling-laws/

6/17Wed
  1. Google Research30

    Google Earth AI releases vectorized dataset mapping hedgerows and woodlands in the British countryside

    Google Research has released a vectorized dataset that converts the high-resolution raster maps in Farmscapes 2020 into actionable inventories of hedgerows, stone walls, and scrubland, covering more than 130,000 square kilometers of the UK. The framework fine-tunes an RSF Vision Transformer backbone pretrained on over 300 million global satellite images and uses Polsby–Popper compactness scores to distinguish woodland, woody patches, and linear woody features.

6/4Thu
5/27Wed
  1. Mistral AI62

    Mistral introduces physical AI, integrating Emmi AI into its enterprise platform

    Mistral says it is bringing Emmi AI into the company and launching physical-AI capabilities for industrial engineering as a new foundational feature of its enterprise solutions. The models predict physical fields directly from geometry and boundary conditions in a single forward pass taking seconds on one GPU, enabling design iteration. Traditional solvers remain available for verification and edge cases.

    Why it matters: Mistral is integrating physics-simulation models into its enterprise platform, offering a view of potential changes to compute requirements and iteration speed in industrial design.

  2. Mistral AI38

    After acquiring Emmi AI, Mistral advances physical-AI research with industrial-simulation results

    After acquiring Emmi AI, Mistral is building foundational physical AI for aerospace, automotive, semiconductors, and energy. Its announced results include AB-UPT, which processes raw geometry with 9M surfaces and 140M volume cells on a single GPU without remeshing, and NeuralDEM, an end-to-end deep-learning surrogate for large multiphysics processes.

5/23Sat
  1. Mistral AI58

    Mistral acquires physics-AI company Emmi AI

    Mistral AI has signed a definitive agreement to acquire Emmi AI, a physics-AI company, to strengthen AI-transformation services for industrial enterprises. Headquartered in Austria, Emmi AI specializes in physical AI and large engineering models that can replace multi-day computations with real-time simulations and build digital twins. Its co-founders and more than 30 researchers and engineers will join Mistral's Science and Applied AI teams in May.

5/20Wed
5/19Tue
  1. Google DeepMind60

    Google DeepMind's Co-Scientist accelerates discovery of genetic clues to reversing cellular aging

    Google DeepMind biologists Omar Abudayyeh and Jonathan Gootenberg used Co-Scientist to support aging research. After scanning tens of thousands of papers, the tool proposed more than 20 new candidate genetic factors. Lab tests validated several hypotheses, successfully making cells younger and improving their overall function. Co-Scientist also analyzed large-scale genetic-screening data, reducing data and literature linkage work that could take up to six months to just days.

    Why it matters: The source reports specific outputs and time comparisons from Co-Scientist's role in aging research, showing how AI agents contribute to scientific hypothesis generation.