🤖 AI 资讯日报 — 2026-06-23 周二
自动生成于 2026-06-23 08:32 | 数据源: T1 官方 / T1.5 媒体 / T2 社区
🏆 T1 官方一手（最高权重）
📌 OpenAI
📌 Anthropic
📌 HuggingFace
📌 GitHub
- AI credits consumed per user now in the Copilot usage metrics API
- The Copilot usage metrics API now reports how many AI credits each user consumed per day, derived from the same AI credits consumption data
- Getting more from each token: How Copilot improves context handling and model routing
- How GitHub Copilot is making more of each session go toward useful work, so your credits go further.
- What are git worktrees, and why should I use them?
- Git worktrees have been around since 2015, but it wasn't until recently they became popular. Learn what they are, how to use them, and why
- MAI-Code-1-Flash available on more Copilot surfaces
- MAI‑Code‑1‑Flash, Microsoft's purpose‑built small coding model, is now available across additional GitHub Copilot surfaces.
- How we built an internal data analytics agent
- Learn how GitHub built Qubot, our internal Copilot-powered analytics agent, to allow any GitHub employee to ask questions about our data in
📌 Apple ML Research
📌 NVIDIA Blog
📌 xAI
📌 Simon Willison
📰 T1.5 媒体报道
🔬 T2 社区 & 学术
📚 Twitter KOL
📚 ArXiv
- From AGI to ASI
- Tim Genewein, Matija Franklin, Alexander Lerchner, Laurent Orseau - Google DeepMind. Published: Jun 09, 2026.
- Patterns, Predictions, and Actions: A Story about Machine Learning
- This graduate textbook on machine learning tells a story of how patterns in data support predictions and consequential actions.
- Optimal Transport for Machine Learners
- Optimal transport is useful because it compares objects by asking how mass should move. Published: May 10, 2025.
- Foundations and Trends in Multimodal Machine Learning: Principles, Challenges, and Open Questions
- This paper provides an overview of the computational and theoretical foundations of multimodal machine learning. Published: Sep 07, 2022.
- Machine Learning Methods for Studying Latent Neural Activity Dynamics
- Recent developments in brain recording are driving demand for machine learning tools capable of decoding latent structure. Published: Jun 09, 2026.
- MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery
- Large language model (LLM) agents are used for automated machine learning algorithm discovery. Published: Jun 09, 2026.
- Machine Learning in Biomechanics: Key Applications and Limitations in Walking, Running, and Sports Movements
- Overview of recent ML applications: pose estimation, feature estimation, event detection. Published: Mar 05, 2025.
- Position: The AI and Machine Learning Community Should Adopt a More Transparent and Regulated Peer Review Process
- Position paper advocating for more transparent, open, and well-regulated peer review in the AI/ML community. Published: Feb 02, 2025.
📚 HF Daily Papers
- When, Where, and How: Adaptive Binning for Tabular Self-Supervised Learning ⬆️1
- Medical tabular data are ubiquitous in clinical research, but deep learning for tables remains underexplored because reliable labels often require costly expert adjudication, even though structured cl...
- Characterizing Narrative Content in Web-scale LLM Pretraining Data ⬆️2
- The narrative composition of web-scale LLM pretraining corpora remains largely unexplored even though narrative is a fundamental mode of human communication. We present the first fine-grained study of...
- SproutRAG: Attention-Guided Tree Search with Progressive Embeddings for Long-Document RAG ⬆️8
- Retrieval-augmented generation (RAG) systems must balance retrieval granularity with contextual coherence, a challenge that existing methods address through LLM-guided chunking, single-level context e...
- MCompassRAG: Topic Metadata as a Semantic Compass for Paragraph-Level Retrieval ⬆️10
- Retrieval-augmented generation (RAG) systems depend critically on how documents are chunked and searched. Fine-grained chunks can improve retrieval precision but expand the search space, increasing la...
- StylisticBias: A Few Human Visual Cues Drive Most Social Biases in MLLMs ⬆️2
- Multimodal large language models (MLLMs) are increasingly deployed in personally and societally consequential settings, yet the visual cues that shape how these models judge people remain poorly under...
- MemSlides: A Hierarchical Memory Driven Agent Framework for Personalized Slide Generation with Multi-turn Local Revision ⬆️14
- Personalized presentation generation requires more than conditioning on a current prompt or template: agents must preserve stable user preferences across tasks, retain newly introduced preferences and...
- Multi-Turn Reflective Masking Elicits Reasoning in Mask Diffusion Models ⬆️9
- While reasoning on autoregressive (AR) models is often performed by chain-of-thought reasoning and reflection, their refinement of previous outputs still relies on fully sequential generation, even wh...
- GeneralVLA-2: Geometry-Aware Reconstruction and Governed Memory for Robot Planning ⬆️3
- Generalist vision-language-action systems need object-centric 3D evidence and reusable manipulation experience to plan reliable robot trajectories. GeneralVLA provides a hierarchical interface for con...
- SpatialAvatar-0: High-Quality 4D Head Avatar with Multi-Stage Reconstruction ⬆️3
- High-quality 4D head avatars from one or a few source portraits are central to telepresence, AR/VR, and digital-human interaction. 3D Gaussian Splatting (3DGS) has emerged as the dominant representati...
- Distilling Examples into Task Instructions: Enhanced In-Context Learning for Real-World B2B Conversations ⬆️2
- In-context learning (ICL) is the standard method for low-resource classification, yet its efficacy in specialized domains remains largely unexplored. We address the challenge of classifying semantical...
- GateMem: Benchmarking Memory Governance in Multi-Principal Shared-Memory Agents ⬆️13
- Memory benchmarks for LLM agents largely assume single-user settings, leaving shared assistants for hospitals, workplaces, campuses, and households understudied. In these deployments, multiple princip...
- BrainG3N: A Dual-Purpose Tokenizer for Controllable 3D Brain MRI Generation ⬆️7
- Three-dimensional (3D) brain MRI is central to clinical neurology and neuro-oncology, where generative models could augment under-represented cohorts, simulate disease trajectories, and support privac...
- WorldLines: Benchmarking and Modeling Long-Horizon Stateful Embodied Agents ⬆️3
- To assist humans over extended periods in real homes, embodied agents must remember user routines, world states, and past interactions. Existing long-term memory benchmarks mainly evaluate language-ce...
- PerceptionDLM: Parallel Region Perception with Multimodal Diffusion Language Models ⬆️50
- Multimodal large language models (MLLMs) have achieved remarkable progress in visual understanding tasks. However, most existing MLLMs rely on autoregressive generation, which limits their efficiency ...
- LedgerAgent: Structured State for Policy-Adherent Tool-Calling Agents ⬆️6
- Policy-adherent tool-calling agents in customer-service domains must maintain task states across turns while calling tools and obeying domain policies. Task states consist of relevant facts, identifie...
本报告由 Hermes Agent 自动生成 | 2026-06-23