00:50 <Vega> OpenAI News | Sullivan & Cromwell brings legal expertise into agreement review 1d ago In collaboration with OpenAI, Sullivan & Cromwell developed Agreement Analyzer, an AI tool for identifying issues in a draft agreement. Read full story

02:39 <Ada> Hugging Face | Scaling Indic Pretraining: 2.8 Trillion Tokens and an Open Data Generation Pipeline Will Jennings · 54m ago · 12 min read A Blog post by NVIDIA on Hugging Face Read full story

04:00 <Ada> arXiv.org | Co-Evolving Acting and Forecasting for Language Agents 4h ago · 3 min read Training language agents is bottlenecked by the high cost of long-horizon rollouts, yet standard agentic reinforcement learning (RL) learns only from a terminal scalar reward, leaving the supervision in intermediate steps largely unused. Prior work exploits this supervision through value targets that estimate per-step success probabilities or observation ... Read full story

04:00 <Ada> arXiv.org | Scaling Multilingual Mixture-of-Experts: How Language Coverage Changes Capacity and Compute Allocation 4h ago · 3 min read Multilingual models share capacity across languages, so adding languages can hurt performance on existing ones. Mixture-of-experts models expand this capacity with little additional per-token computation. Given this extra capacity, does adding experts reduce what each additional language costs? To answer this question, we propose a repetition-aware joint ... Read full story

04:00 <Ada> arXiv.org | ReTerm: Recovery Training for Self-Improving Terminal Agents 4h ago · 3 min read Terminal agents have shown strong abilities in complex coding tasks. However, they often struggle to recover from errors during long-running execution (e.g., command errors, faulty implementations), eventually leading to task failures. To this end, we present ReTerm, a self-improving terminal agent trained by learning how to recover from its own failed te... Read full story

04:00 <Ada> arXiv.org | SPARCL: Spatially Adaptive Regional Coverage for Active Learning in Chest X-ray Analysis 4h ago · 3 min read Pixel-level annotation for chest X-ray (CXR) lesion segmentation is expensive, making active learning (AL) essential. However, existing AL strategies fail to account for the spatial distribution of selected samples, resulting in inconsistent performance across focal (e.g., aortic calcification) and diffuse (e.g., nodules) pathologies. In this paper, we pr... Read full story

04:00 <Ada> arXiv.org | Morphology-aware deep learning predicts colorectal cancer survival from whole-slide images 4h ago · 3 min read Background: Five-year survival prediction from hematoxylin-and-eosin whole-slide images could support risk stratification, but the task is weakly supervised, and the tissue morphologies carrying prognostic information remain incompletely characterized. Methods: We conducted a retrospective secondary analysis using TCGA-COAD/READ (n=562) for morphology-spe... Read full story

04:00 <Ada> arXiv.org | A Bethe-Ansatz Source of Early Token Correlations: A marginal-preserving construction for parallel proposals Abraham Chachamovits · 4h ago · 3 min read Parallel token proposals need a joint law, not merely accurate token marginals. We construct coordinate-Bethe-ansatz laws from the periodic XXZ chain and compare them with their maximum-entropy pairwise projections at identical one- and two-site moments. In an eight-site, three-excitation sector, symmetry makes the full probability-law difference one-dime... Read full story

04:00 <Ada> arXiv.org | PMB-NN: Physiological-Model-Based Neural Network for Glucose Forecasting in Type 1 Diabetes Yaowen Zhang, Shuhao Que, Ying Wang · 4h ago · 3 min read Accurate, physiologically consistent glucose forecasts could help people with type 1 diabetes respond to impending excursions. We applied our physiological-model-based neural network (PMB--NN) framework to glucose forecasting. A temporal convolutional network (TCN) used 180 min of history, constrained during training by a personalized four-parameter physi... Read full story

04:01 <Ada> arXiv.org | AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery 4h ago · 3 min read Formulaic alpha mining seeks to discover executable expressions that capture predictive market signals in a vast, noisy search space. Recent LLM-based approaches use prompting, iterative search, or seed-factor evolution, leaving direct factor construction through policy learning less explored. To address this, we introduce AlphaRLM, a reinforcement-learni... Read full story

04:01 <Ada> arXiv.org | Self-improving data synthesis enables scalable adaptation of embedding models Zucan Lyu, Rui Tang, Jianlyu Chen, Peisheng Jin, Defu Lian · 4h ago · 3 min read AI agents require specialised retrieval across domains and tasks, but adapting embedding models is constrained by costly manual training-data design. Here we introduce SYNTRA, a self-improving agentic framework that turns a few retrieval examples into reusable synthesis instructions. Given an unlabelled corpus and a few relevant query-document pairs, SYNT... Read full story

04:01 <Ada> arXiv.org | EV-JEPA: Elastic V-JEPA for Any Device and Cross-Device Inference 4h ago · 3 min read Foundation models like V-JEPA are increasingly deployed across a heterogeneous edge-cloud continuum spanning smart glasses, phones, robots, and servers. Yet, they are typically trained as rigid, fixed-capacity networks bound to one device, supporting neither reconfigurable compute budgets nor collaborative execution. Existing approaches address only one a... Read full story

04:51 <Ada> arXiv.org | Hindsight-Privileged RLVR: Single-Rollout Token Credit Without Group Sampling 4h ago · 3 min read Reinforcement learning with verifiable rewards (RLVR) trains language models from checked task outcomes, but assigning a terminal reward to individual reasoning tokens or actions remains difficult. Group-relative methods such as Group Relative Policy Optimization (GRPO) sample several attempts per prompt, but lose their reward-based learning signal when a... Read full story

04:51 <Ada> arXiv.org | Analyzing performance of Retrieval Augmented Generation models for Question Answering Tasks 4h ago · 3 min read Retrieval-Augmented Generation (RAG) models have become almost a cornerstone of open-domain question answering tasks; in which the quality of answers generated in the process, greatly relies on the effectiveness of retrieval mechanisms. In the realm of LLM, the last two years have seen tremendous progress in RAG architectures, with many models reporting t... Read full story

04:51 <Ada> arXiv.org | Retrieval Over Training: Similarity search-based Model Selection for Time Series Anomaly Detection 4h ago · 3 min read Anomaly detection is a fundamental task for time-series analytics, with important implications for the downstream performance of many applications. Despite the large number of anomaly detection methods proposed in the literature, recent benchmark studies have shown that no single detector performs best across highly heterogeneous time series. Therefore, a... Read full story

04:51 <Ada> arXiv.org | SWE-Nexus: Automated Composition of Software Engineering Tasks from Test Execution 4h ago · 3 min read Historical code changes provide convenient task boundaries, but do not identify which production functions selected tests exercise or support composing validated behaviors beyond a single change. We present SWE-Nexus, a fully automated pipeline that constructs repository-level implementation tasks from test execution. Four construction agents select tests... Read full story

04:51 <Ada> arXiv.org | Frame Evidence:A Benchmark for Reference Frames and Visual Evidence in Multi-Image Spatial Reasoning 4h ago · 3 min read A spatial answer depends on the reference frame specified in language and its supporting visual evidence. Item accuracy alone obscures whether correctness persists when either changes. We introduce Frame Evidence, a benchmark of 320 matched triplets and 960 Chinese multiple-choice questions from captured and rendered scenes. One family compares three grou... Read full story

04:51 <Ada> arXiv.org | Why Do Agent Teams Fail Across Boundaries? A Failure Taxonomy for Agent-to-Agent Systems with Private States 4h ago · 4 min read Agents are increasingly interacting with other agents that act for different principals (people or organizations on whose behalf they work), such as a buyer's personal agent negotiating with an enterprise's seller agent. In such agent-to-agent (A2A) scenarios, each agent has its own environment and holds information and authority that the other cannot see... Read full story

04:51 <Ada> arXiv.org | DexMimic: A Generalist Dexterous Hand-Object Tracking Policy via Value-Guided Reset Allocation 4h ago · 3 min read Achieving human-like dexterous manipulation remains a central challenge in robotics. Learning from human hand-object interaction (HOI) datasets offers a promising pathway to diverse dexterous skills, yet existing physics-based approaches train specialist policies for individual trajectories, thereby limiting skill reuse at scale. We introduce DexMimic, a ... Read full story

04:51 <Ada> arXiv.org | Where Do LLM Ensemble Gains Come From? Decomposing Candidate Value, Diversity, and Aggregation Donghwan Kim · 4h ago · 3 min read LLM ensemble gains are often measured against a single baseline, but the value attributed to additional candidates can change with the comparison used. We study an LLM aggregator under four matched conditions: Direct (no candidates), Single-slot (one candidate in a multi-slot interface), Duplicate (one candidate repeated), and Multi-candidate (separately ... Read full story

04:51 <Ada> arXiv.org | Knowledge Graphs as Files: An Agent Harness Where Tool Creation and Tool Use Co-Evolve Yuyang Bai, Yu Wang, Yu Zhang · 4h ago · 3 min read Searching text-rich knowledge graphs requires following relations while examining textual evidence. Predefined graph interfaces constrain what evidence agents can access, even when they write code to combine tool calls or process retrieved subgraphs. We propose direct knowledge graph interaction (DKGI), a harness in which language model agents create grap... Read full story

04:51 <Ada> arXiv.org | When Do Learned Blocks Preserve Causal Meaning? Block Identifiability and Constructive Causal Abstractions Karim Zaghw, Wieland Brendel, Patrik Reizinger · 4h ago · 3 min read Standard identifiability theory assumes one-dimensional latent factors, but learned representations can encode latent information across multidimensional blocks whose internal coordinates are not uniquely determined. When such blocks are used to represent and intervene on higher-level concepts, it is unclear whether their causal meaning survives the trans... Read full story

14:04 <Ada> Hugging Face | Welcome Beam: a 501B open model you can own, customize, and scale ben burtenshaw · 14h ago · 21 min read We’re on a journey to advance and democratize artificial intelligence through open source and open science. Read full story

17:03 <Vega> MIT News | Massachusetts Institute of Technology | Fostering community amid Kendall Square’s innovation Zach Winn | MIT News · 5h ago · 6 min read The 18th annual Kendall Square Association Annual Meeting explored how new technologies could be a force for shared prosperity in society. Read full story

19:04 <Vega> The latest research from Google | Grounding AI agents in biological reality: Building blocks for autonomous wet labs 52m ago · 10 min read General Science Read full story

20:35 <Vega> OpenAI News | Stack Overflow for Agents triples signups with ChatGPT plugin 20h ago Two engineers shipped Stack Overflow for Agents in a few weeks with Codex, and its ChatGPT plugin helped drive up to 10× growth in agent creation. Read full story

23:36 <Vega> OpenAI News | How Itaú is building generative AI for its entire customer base 4h ago With ia.i, a conversational interface, and the IARA governance platform, Itaú is applying its AI First strategy to customers and employees — in one of Brazil’s most regulated sectors. Read full story

#ai за 2026-10-16