The Archive

LLM Chat Completions Server 0.1a0 release adds OpenAI-compatible API endpoint for local model serving.

Meta says AI is making it dramatically easier to build and launch new consumer apps, with CEO Mark Zuckerberg telling investors the company has more new consumer products on the way following a recent wave of releases for Facebook Groups, Marketplace sellers, Instagram, and gaming.

LLM 0.32rc1 introduces content-addressable message storage and schema redesign enabling conversation branching.

British AI neocloud Nscale is buying software startup Anyscale, which helps companies scale their AI workloads across data centers and servers.

Federated learning method using encrypted metadata clustering to balance privacy, communication, and computation without sacrificing one dimension.

Perception-Correction Distillation isolates perception failures in multimodal reasoners using teacher-student disagreement without labeled ground truth.

Google DeepMind releases Gemini Robotics ER 2 with video understanding, task orchestration, and multi-robot coordination capabilities.

Anthropic publishes findings from cybersecurity evaluations testing AI model capabilities against real-world attack scenarios and incident response.

A new study estimates only 2,000 U.S. engineers have the expertise to deliver meaningful AI ROI, as enterprises race to hire forward-deployed engineers to implement AI at scale.

CARP reputation-penalty mechanism prevents LLM agents from fabricating product listings using complaint signals without access to ground truth.

Gap Index metric evaluates dimensionality reduction scatterplot quality by measuring distortion in empty regions alongside traditional point-distance metrics.

Fairness Pruning localizes demographic bias in GLU-MLP layers via differential neuron activations, enabling targeted bias mitigation in LLMs.

PathView-Bench evaluates multimodal LLMs on fine-grained, multiscale pathology image understanding with spatial annotations across 23 datasets.

Plus, a new deprecation policy ensures features aren't removed suddenly.

Budget-constrained human audit allocation for N LLM agents identifies miscalibration threshold where confidence-ranking underperforms random selection.

Cybersecurity experts told TechCrunch that one of the biggest lessons to be taken from the OpenAI hack against HuggingFace has nothing to do with AI, but traditional cybersecurity defense.

ObjectStream organizes streaming video context around persistent latent objects as memory anchors instead of token or temporal importance.

FICE enables fully inductive cardinality estimation for SPARQL queries over unseen knowledge graphs without retraining.

Expert Subspace Separation Index disentangles route coherence from candidate quality in sparse MoE routing using controlled interventions.

Empirical study of LLM ability to synthesize microservice architectures from natural-language requirements including service definitions and interactions.

Marlinks-NS dataset of submarine distributed acoustic sensing and AIS data for vessel detection and cable-protection machine-learning tasks.

Ensemble consensus semi-supervised learning for molecular property prediction without label-preserving augmentations.

HARGO framework applies heterogeneity-aware RL post-training to optimize LLM behavior on HPC tasks like data-race detection beyond SFT accuracy.

MonoVoc decouples geometry and semantics for lightweight monocular open-vocabulary 3D Gaussian scene understanding via training-free pipeline.

Hardware-aware NAS and fine-tuning for RGB-D affordance segmentation on embedded wearable robots via Pareto-optimal architecture search.

CACHE-UK memory editor for sequential fact updates in 4-bit quantized LLMs deployed in finance, addressing quantization stability crisis.

Tycho coding-agent constructs programmatic world models for ARC-AGI-3, inferring game rules and hidden state via structured history.

Novel LLM evaluation framework using pairwise comparisons across multiple models to approximate expert-level assessment without reference standards.

MORFES benchmark: 500 expert-verified items testing inflectional morphology recognition and production for Modern Greek language models.