Data & training

AIBerkeley AI Research authors argue that near-free inference will make agents the dominant workload for data systems, requiring redesign for agentic speculation, agent-run state and coordination, and agent-synthesized systems.

Why it matters: The piece maps three concrete data-system challenges posed by near-free inference, useful for anyone designing infrastructure for agent workloads and memory.

AIMeta released Brain2Qwerty v2, an end-to-end deep learning pipeline that decodes sentences in real time from non-invasive brain recordings.

Why it matters: The source reports word accuracy and data-scaling results for non-invasive decoding, offering a benchmark against surgical brain-computer interfaces and prior non-invasive methods.

AIOpenAI reports that reinforcement learning on realistic conversations targeting traits such as honesty, epistemic humility, and corrigibility improved a model across 44 out-of-distribution alignment evaluations.

Why it matters: The post tests whether reinforcement learning on beneficial traits in one domain transfers to unrelated alignment benchmarks and holds up under adversarial steering.

AIOpenAI's alignment team found that re-generating 100,000 WildChat conversations with five recent OpenAI models predicted production failure rates across four orders of magnitude, with 95% of predictions within 1.04 orders of magnitude.

Why it matters: The post tests whether public chat data can predict real production failure rates, and where that prediction breaks down for agentic behavior.

AIXiaomi MiMo and the TileRT team say a 1T model exceeds 1,000 tps on a single standard 8-GPU node using general-purpose GPUs.

Why it matters: The post shows how FP4 quantization, DFlash speculative decoding, and TileRT kernels combine to reach over 1,000 tps on a standard GPU node.

AIMiniMax says its M2 series could not output the name Ma Jiaqi, a failure it traced to post-training data that rarely included the token.

Why it matters: The post traces a community-noticed token failure through tokenizer, embedding, and lm_head tests, showing how post-training data coverage can cause low-frequency token drift.

AIStability AI released Stable Audio 3.0, a family of four audio models trained on fully licensed data.

Why it matters: The source specifies which models are open-weight, their licensing terms, and clip-length limits, which matters for anyone deciding whether to build on them.

AICognition and Applied Compute RL-trained SWE-check, a specialized bug detection model for the Windsurf IDE. It matches frontier performance on in-distribution evals and is an order of magnitude faster with cheaper inference, though it trails frontier models on out-of-distribution evals (delta F1 0.29 versus 0.49 before training).

Why it matters: The post explains how production environment replication, reward linearization, and two-phase post-training trade bug-detection quality against latency for an IDE specialist model.

AIWerner Vogels announced S3 Files, which lets users mount any S3 bucket as a filesystem without making copies, running sync scripts, or choosing between file and object storage.

Why it matters: The post explains how S3 Files changes access to existing buckets, which matters for teams weighing file and object storage workflows.

AITsinghua AIR and Xiaomi Auto released Hyper Diffusion Planner (HDP), a diffusion-based end-to-end autonomous driving planner tested on real vehicles.

Why it matters: The report shows how diffusion planning was tuned for real road driving, which helps readers judge what changes beyond open-loop benchmarks.

AITsinghua AIR and ByteDance Seed introduce CUDA Agent, a large-scale agentic reinforcement learning system for generating high-performance CUDA kernels.

Why it matters: The source explains how the CUDA Agent system uses agentic reinforcement learning and a staged training recipe to generate GPU kernels faster than torch.compile.

AIResearchers trained a humanoid with 22-DoF dexterous hands mainly on over 20,000 hours of egocentric human video, with no robot in the loop, to perform tasks such as assembling model cars and folding shirts.

Why it matters: The source reports a measured scaling law linking human video volume to robot action loss, which bears on how dexterous robot training data might be collected and reused.

AITsinghua AIR researchers published PocketXMol in Cell, a generative AI model that uses atomic-level task prompts to handle 13 protein-pocket interaction tasks without task-specific fine-tuning.

Why it matters: The paper unifies several molecular generation tasks in one model and reports wet-lab validation of designed inhibitors and peptides, useful for comparing AI drug discovery approaches.

AIJim Fan announced DreamDojo, an open-source interactive world model that takes robot motor controls and generates future frames in pixels.

Why it matters: The post explains how human egocentric videos are converted into latent actions, a method that reduces dependence on robot-collected data for training world models.

AIMiniMax describes Forge, its internal reinforcement learning framework for training real-world agents, which was used during the development of MiniMax M2.5.

Why it matters: The post details how the Forge framework balances throughput, stability, and agent flexibility, with concrete scheduling and prefix-merging methods for training agent RL at scale.

AIAnthropic reports that resource configuration alone can move Terminal-Bench 2.0 scores by up to 6 percentage points, with infra error rates falling from 5.8% under strict enforcement to 0.5% when uncapped.

Why it matters: The source shows how container resource limits shift agentic coding scores, which helps readers interpret small leaderboard gaps and set up evals more consistently.

AIIntern Large Models introduces Intern-S1-Pro, a 1T-parameter MoE open-source multimodal scientific reasoning model with 1T-A22B active configuration.

Why it matters: The post pairs a 1T MoE open-source scientific reasoning model with benchmark tables against closed models, letting readers compare claimed strengths directly.

AIZ.ai has released GLM-OCR, a 0.9B-parameter multimodal OCR model for complex document understanding, under the MIT License.

Why it matters: The page gives benchmark scores, a 0.9B parameter size, and supported serving frameworks, which help readers weigh OCR deployment options against heavier alternatives.

AITim Dettmers describes building SERA, Ai2's first Open Coding Agents release, using 32 GPUs and synthetic data.

Why it matters: The post explains how a small team built an open coding agent with cheap synthetic data and soft verification, a reusable recipe for specializing models on private code.

AIThinking Machines Lab describes on-policy distillation, which samples rollouts from a student model and has a teacher grade each token with reverse KL. The authors report that this matches Qwen3-style reasoning results at a fraction of RL's cost, with AIME'24 reaching 70% in about 150 steps from a 400k SFT checkpoint.

Why it matters: The post explains why on-policy distillation gives dense per-token feedback, letting a small model match RL results at much lower compute cost.