5arXiv cs.AI (Artificial Intelligence)·
GigaPath-Flash and GigaTIME-Flash: Efficient open-weight pathology foundation models for whole-slide analysis
Microsoft Research introduces GigaPath-Flash and GigaTIME-Flash, compact pathology foundation models that retain 97% of the billion-parameter GigaPath's slide-level performance at 50x lower compute via knowledge distillation into a 22M-parameter ViT-S tile encoder. GigaTIME-Flash extends the backbone to predict tumor immune microenvironment from H&E images, outperforming the original CNN-based GigaTIME while running 6x faster and using 8x less GPU memory. Both models are released as open weights under Apache 2.0, addressing key barriers of compute cost and licensing that have limited clinical deployment of pathology AI.
Related guides (2)
Related events (8)
Atlas H&E-TME: AI system matches expert pathologist accuracy for scalable tumor microenvironment profiling
Researchers present Atlas H&E-TME, an AI system built on the Atlas family of pathology foundation models that generates over 4,500 quantitative readouts per whole-slide H&E image at cell-level resolution across multiple cancer types. The system is validated using a novel dual framework: an IHC-informed multi-pathologist consensus protocol for depth, and benchmarking against 200,000+ annotations across 1,500+ cases from 25+ sources spanning eight cancer types. Atlas H&E-TME matches or exceeds pathologist H&E-only performance, demonstrating that standard histopathology slides can serve as a scalable quantitative window into the tumor microenvironment. The work advances computational pathology by enabling tissue-based biomarker discovery without requiring specialized staining modalities.
DeepMind Launches 27B Parameter Gemma-Based Foundation Model for Single-Cell Analysis
DeepMind has released a new 27 billion parameter foundation model built on the Gemma open-model family, specifically designed for single-cell biological analysis. The model contributed to the discovery of a new potential cancer therapy pathway. This represents a significant application of large language model architecture to computational biology and genomics research.
Qwen3.5 Small tops mobile-sized open models; GPT-5.3 Instant, Gemini 3.1 Flash-Lite, Claude memory import, and LLM deanonymization research
Alibaba released the Qwen3.5 Small model series (0.8B–9B parameters) with a hybrid Gated Delta Networks + sparse MoE architecture, with the 9B model outperforming OpenAI's gpt-oss-120B on GPQA Diamond despite being 13.5x smaller; all weights are Apache 2.0 licensed. Google introduced Gemini 3.1 Flash-Lite, a cost-optimized model at $0.25/M input tokens with 2.5x faster TTFT than Gemini 2.5 Flash. OpenAI released GPT-5.3 Instant targeting conversational quality improvements and hallucination reduction, while Anthropic added memory import/export functionality across all Claude tiers. Separately, researchers from MATS, Anthropic, and ETH Zurich demonstrated that LLM-based pipelines can deanonymize pseudonymous online users at 68% recall/90% precision for $1–4 per profile.
DeepSeek V4 Preview Release: 1.6T-param Pro and 284B Flash Models with 1M Context, Open-Sourced
DeepSeek has released DeepSeek-V4 as an open-weights preview, comprising two MoE variants: V4-Pro (1.6T total / 49B active parameters) and V4-Flash (284B total / 13B active parameters). Both models support 1M token context by default, enabled by a novel Token-wise compression and DeepSeek Sparse Attention (DSA) architecture. V4-Pro claims open-source SOTA on agentic coding benchmarks and world-class math/STEM/coding performance rivaling top closed-source models, while V4-Flash offers near-parity reasoning at lower cost and latency. The API is live today with OpenAI and Anthropic compatibility, and legacy model endpoints will be retired in July 2026.
GPIC: Stanford Releases 28-Trillion-Pixel Permissively Licensed Image Corpus for Visual Generation Research
Stanford Vision Lab introduces GPIC, a Giant Permissive Image Corpus of approximately 28 trillion pixels comprising 100M training, 200K validation, and 1M test images, all permissively licensed for research and commercial use. Images are captioned by a state-of-the-art vision-language model, safety-filtered, deduplicated, and hosted on Hugging Face. The release includes a benchmarking protocol for generative modeling and a reference baseline using pixel-space flow matching. The dataset addresses a key gap in scalable visual generative modeling research by providing a large, stable, and openly licensed resource.
Accelerating Vision-Language Models: BridgeTower on Habana Gaudi2
This Hugging Face blog post covers the deployment and acceleration of BridgeTower, a vision-language model, on Intel's Habana Gaudi2 AI accelerator hardware. The piece likely benchmarks inference throughput and training performance on Gaudi2 compared to other hardware. It represents a practical infrastructure and deployment case study for multimodal models on alternative AI accelerators.
MambaGaze: Bidirectional Mamba with Explicit Missing Data Modeling for Cognitive Load Assessment from Eye-Gaze Tracking
MambaGaze is a framework for real-time cognitive load assessment from eye-tracking data, combining XMD encoding (observation masks and time-deltas for missing data) with bidirectional Mamba-2 for efficient long-range temporal modeling. Evaluated on CLARE and CL-Drive datasets under leave-one-subject-out protocol, it achieves 76.8% and 73.1% accuracy, outperforming CNN, Transformer, ResNet, and VGG baselines by 4-12 percentage points. Edge deployment on NVIDIA Jetson platforms achieves 43-68 FPS at under 7.5W, demonstrating feasibility for wearable and safety-critical applications such as driver vigilance monitoring.
The Batch: Jalapeño inference chip, Fugu multi-agent system, Claude Tag, Robin bio-agent, and Getty-OpenAI deal
OpenAI and Broadcom announced Jalapeño, OpenAI's first custom inference chip, designed in nine months with AI-assisted design and showing better performance-per-watt than current accelerators; engineering samples are already running GPT-5.3-Codex-Spark with datacenter deployment planned by end of 2026. Sakana AI released Fugu, a multi-agent routing system that scored 73.7% on SWE-Bench Pro, outperforming Claude Opus 4.8 and GPT-5.5 while remaining below the inaccessible Fable 5. Additional items cover Anthropic's Claude Tag Slack integration for async team collaboration, Seedance 2.5 video model improvements, the Robin autonomous biology research agent that identified a novel drug candidate, and a Getty Images licensing partnership with OpenAI.