AllLLMRoboticsStartupsResearchEthicsBusinessHardwareAutonomousMedicineCreative

All news

7718
GRAPHEVAL: New Graph Framework Reveals That Self-Consistency Masks Reasoning Errors in LLMs
LLM·arXiv cs.CL

GRAPHEVAL: New Graph Framework Reveals That Self-Consistency Masks Reasoning Errors in LLMs

Jul 10, 2026·2 min
Self-Evolving LLM Agents: Compiling SOPs into Tools Reduces Latency by 42%
LLM

Self-Evolving LLM Agents: Compiling SOPs into Tools Reduces Latency by 42%

Jul 10, 2026·2 min
Graph neural network recognizes hand gestures in 48 ms: 99% accuracy for prosthetics and AR
LLM

Graph neural network recognizes hand gestures in 48 ms: 99% accuracy for prosthetics and AR

Jul 10, 2026·3 min
Qwen3 language models easily make educational tasks more complex, but struggle to simplify them
LLM

Qwen3 language models easily make educational tasks more complex, but struggle to simplify them

Jul 10, 2026·2 min
NYIT developed an LLM assistant for an applied analysis course in MATLAB
LLM·arXiv cs.AI

NYIT developed an LLM assistant for an applied analysis course in MATLAB

Jul 10, 2026·2 min
Infinity-Parser2 surpassed DeepSeek-OCR-2 in document parsing and opened a dataset of 5 million examples
LLM

Infinity-Parser2 surpassed DeepSeek-OCR-2 in document parsing and opened a dataset of 5 million examples

Jul 10, 2026·3 min
Hallucination Self-Play: a small language model surpassed large LLMs in hallucination detection
LLM

Hallucination Self-Play: a small language model surpassed large LLMs in hallucination detection

Jul 10, 2026·2 min
Gemini 2.5 Flash replaces human evaluators in voice agent testing
LLM·arXiv cs.CL

Gemini 2.5 Flash replaces human evaluators in voice agent testing

Researchers tested how reliably Gemini can evaluate voice AI agents instead of humans — and found that the model performs at 100 times lower

Jul 10, 2026·3 min
Top themes
01Idiobionics: Scientists Warn of the Risks of Hacking Smart Bionic ProsthesesJul 10, 2026·2 min02TACO: a method removes credit “contamination” in RL training of language modelsJul 10, 2026·2 min03New MiniLM-based method more accurately detects queries outside chatbots' knowledge scopeJul 10, 2026·2 min04Scientists propose "alignment plausibility" as a safety standard for medical LLMsJul 10, 2026·3 min05Debiasing NLP models creates new stereotypes for other demographic groupsJul 10, 2026·3 min
Review of 18 medical LLMs: narrow specialists diagnose more accurately than general-purpose models
LLM·arXiv cs.AI

Review of 18 medical LLMs: narrow specialists diagnose more accurately than general-purpose models

Researchers compared 18 language models across a five-level scale of medical competencies — and found that specialized and general-purpose A

Jul 10, 2026·2 min
01EspanStereo: new dataset reveals LLM stereotypes in Spanish-speaking culturesJul 10, 2026·2 min02DeepSearch-Evolve: a 9B-parameter web agent learns without supervision and scores 61% on GAIAJul 10, 2026·2 min03Researchers described how LLM agents and humans undermine trust in information chainsJul 10, 2026·2 min
LLMs prove theorems but do not discover mathematics: researchers outline the next step
LLM·arXiv cs.CL

LLMs prove theorems but do not discover mathematics: researchers outline the next step

A position paper on arXiv systematizes progress in AI4Math and explains why current LLM systems are incapable of discovering new theorems —

Jul 10, 2026·2 min
New AI tool combines GTAP and APSIM to assess shocks in agricultural supply chains
LLM·arXiv cs.AI

New AI tool combines GTAP and APSIM to assess shocks in agricultural supply chains

Researchers have developed an AI system based on GTAP and APSIM that analyzes shocks in agricultural supply chains through natural-language

Jul 10, 2026·3 min
LSTM outperformed Random Forest and Naive Bayes in tweet sentiment analysis with a ROC-AUC of 0.92
LLM·arXiv cs.CL

LSTM outperformed Random Forest and Naive Bayes in tweet sentiment analysis with a ROC-AUC of 0.92

Researchers on arXiv compared five classical ML algorithms and an LSTM neural network on a Kaggle Twitter dataset: the recurrent network achieved 80% accuracy and a ROC-AUC of 0.92, outperforming all

Jul 10, 2026·3 min
Context Graph: researchers taught enterprise AI agents to work without queries
LLM·arXiv cs.AI

Context Graph: researchers taught enterprise AI agents to work without queries

Jul 10, 2026·2 min
LiST: a new method simultaneously delivers accuracy, robustness, and calibration in neural networks
LLM·arXiv cs.LG

LiST: a new method simultaneously delivers accuracy, robustness, and calibration in neural networks

Jul 10, 2026·2 min
Hol-PCFG cut syntactic parsing parameters by 99.94% and took the lead in six languages
LLM·arXiv cs.CL

Hol-PCFG cut syntactic parsing parameters by 99.94% and took the lead in six languages

Jul 10, 2026·2 min
MetaNCA teaches neural networks to self-organize without backpropagation
LLM·arXiv cs.LG

MetaNCA teaches neural networks to self-organize without backpropagation

Jul 10, 2026·2 min
Jet-Long extends language model context windows to 128K without fine-tuning
LLM·arXiv cs.LG

Jet-Long extends language model context windows to 128K without fine-tuning

Jul 10, 2026·2 min
Eternis-Forecaster 8B: activation probes are more accurate than reasoning chains and save up to 47% of tokens
LLM

Eternis-Forecaster 8B: activation probes are more accurate than reasoning chains and save up to 47% of tokens

Jul 10, 2026·2 min
Datalab Lift vs Competitors: How a 9B Document Extractor Works with JSON Schema
LLM

Datalab Lift vs Competitors: How a 9B Document Extractor Works with JSON Schema

Jul 9, 2026·2 min
Anthropic launches Reflect dashboard to analyze how you use Claude
LLM

Anthropic launches Reflect dashboard to analyze how you use Claude

Jul 9, 2026·3 min
Meta released Muse Spark 1.1 — multimodal agent with 1 million token context
LLM·Meta AI Blog

Meta released Muse Spark 1.1 — multimodal agent with 1 million token context

Meta Superintelligence Labs introduced Muse Spark 1.1 — an agentic model with a 1 million token context window, video and audio support, and

Jul 9, 2026·2 min
Mistral Updates Studio: Version Control System for Enterprise AI Prompts
LLM·Mistral AI News

Mistral Updates Studio: Version Control System for Enterprise AI Prompts

Mistral AI added centralized prompt management to Studio — now every AI instruction has a version, owner, and change log without engineer in

Jul 9, 2026·2 min