Skip to content
gentic.news — AI News Intelligence Platform
Connecting to the Living Graph…
gentic.news

AI Research

Breaking AI research news: latest papers from arXiv, NeurIPS, ICML, and top labs. Track transformer architecture advances, reasoning breakthroughs, and scientific discoveries in machine learning and artificial intelligence.

AI ResearchFunding & BusinessProducts & LaunchesBig TechStartupsOpen SourcePolicy & EthicsOpinion & Analysis
A sleek blue and white AI robot face with glowing eyes against a dark tech background, representing Tencent's…
AI Research
95

Tencent's Training-Free GRPO: A Paradigm Shift in AI Alignment Without Fine-Tuning

Tencent researchers have introduced Training-Free GRPO, a method that achieves reinforcement learning-level alignment results for just $18 instead of $10,000—with zero parameter updates. This breakthrough could fundamentally change how we optimize language models.

twitter.com/Feb 16, 2026/3 min read
natural language processingmachine learningai research
Researchers examining a futuristic AI model visualization on a large screen, with data flow graphics and server…
AI Research
85

New AI Framework Promises to Revolutionize Model Training Efficiency

Researchers have introduced a novel AI training framework that dramatically reduces computational requirements while maintaining performance. This breakthrough could make advanced AI development more accessible and sustainable.

twitter.com/Feb 16, 2026/3 min read
machine learningcomputational efficiencyai research
Side-by-side screenshots of Claude and GPT interfaces displaying code and analysis, with a progress bar labeled…
AI Research
72

AI's Time Horizon Expands: Claude and GPT Push Multi-Hour Task Capabilities

New analysis reveals Claude Opus 4.6 and GPT 5.3 Codex can handle complex tasks requiring hours of human effort. The METR benchmark shows AI systems approaching 3-4 hour time horizons at 50% success rates, signaling major progress in sustained reasoning.

lesswrong.com/Feb 16, 2026/3 min read
software developmentai benchmarksfrontier models
A digital visualization of a vast, interconnected neural network with a glowing 1M token context window, symbolizing…
AI Research
70

Beyond the Token Limit: How Claude Opus 4.6's Architectural Breakthrough Enables True Long-Context Reasoning

Anthropic's Claude Opus 4.6 represents a fundamental shift in large language model architecture, moving beyond simple token expansion to create genuinely autonomous reasoning systems. The breakthrough enables practical use of million-token contexts through novel memory management and hierarchical processing.

pub.towardsai.net/Feb 15, 2026/3 min read
natural language processingmachine learningai research
A doctor examines a medical scan on a monitor while AI-generated text overlays highlight clinical terms, with a…
AI Research
70

Medical AI Breakthrough: New Method Teaches Vision-Language Models to Understand Clinical Negation

Researchers have developed a novel fine-tuning technique that significantly improves how medical vision-language models understand negation in clinical reports. The method uses causal tracing to identify which neural network layers are most responsible for processing negative statements, then selectively trains those layers.

arxiv.org/Feb 13, 2026/3 min read
nlphealthcare-technologymedical-ai

SkillsBench Reveals AI Agent Skills: Powerful But Unpredi…

AI Research
75

SkillsBench Reveals AI Agent Skills: Powerful But Unpredictable

A new benchmark reveals AI agent skills boost performance by 16% on average, but benefits vary wildly across domains. Surprisingly, models can't create the skills they benefit from using.

arxiv.org/Feb 13, 2026/3 min read
machine learningbenchmarksai research
Researchers present a screenshot of BrowseComp-V³, a benchmark interface showing AI models failing at complex visual…
AI Research
75

New Benchmark Exposes Critical Gaps in AI's Ability to Navigate the Visual Web

Researchers unveil BrowseComp-V³, a challenging new benchmark testing multimodal AI's ability to perform deep web searches combining text and images. Even top models score only 36%, revealing fundamental limitations in visual-text integration and complex reasoning.

arxiv.org/Feb 13, 2026/3 min read
natural language processingcomputer visionai research
Two side-by-side dashboard views of a self-driving car navigating a rainy city street at night, with one showing…
AI Research
72

The Unseen Storm: How AI Researchers Are Making Self-Driving Cars See Through Rain and Fog

Researchers have developed a new benchmark called navdream to isolate how weather and lighting affect autonomous driving AI, separate from road complexity. They found current systems degrade significantly in appearance shifts, and propose a universal interface using DINOv3 for robust zero-shot performance.

arxiv.org/Feb 13, 2026/3 min read
autonomous vehiclescomputer visionai research
AI system interface displaying a world map with location pins and geographic data overlays, symbolizing GeoAgent's…
AI Research
70

GeoAgent: AI That Thinks Like a Geographer to Pinpoint Any Location

Researchers unveil GeoAgent, an AI system that masters geolocation by learning from human geographic reasoning. It uses expert-annotated data and novel rewards to ensure its logic aligns with real-world geography, outperforming existing models.

arxiv.org/Feb 13, 2026/3 min read
geospatial analysismachine learningartificial intelligence
A close-up of Cerebras wafer-scale chip hardware with intricate circuitry, reflecting a shift in AI computing…
AI Research
75

Beyond Nvidia: How OpenAI's Cerebras-Powered Model Redefines AI Hardware Competition

OpenAI's GPT-5.3-Codex-Spark demonstrates real-time coding capabilities on Cerebras hardware, challenging Nvidia's dominance and signaling a new era of specialized AI infrastructure.

aibusiness.com/Feb 13, 2026/3 min read
tech innovationcerebrasai hardware
Two data scientists examine visualizations of overlapping word clusters on a monitor, highlighting semantic…
AI Research
70

The Hidden Contamination Crisis: How Semantic Duplicates Are Skewing AI Benchmark Results

New research reveals that LLM training data contains widespread 'soft contamination' through semantic duplicates of benchmark test data, artificially inflating performance metrics and raising questions about genuine AI capability improvements.

arxiv.org/Feb 12, 2026/3 min read
ai ethicsmachine learningresearch methodology
Two AI agents face off in a game-theory matrix, one choosing cooperation, the other defection, with a red warning…
AI Research
75

Game Theory Exposes Critical Gaps in AI Safety: New Benchmark Reveals Multi-Agent Risks

Researchers have developed GT-HarmBench, a groundbreaking benchmark testing AI safety through game theory. The study reveals frontier models choose socially beneficial actions only 62% of time in multi-agent scenarios, highlighting significant coordination risks.

arxiv.org/Feb 12, 2026/3 min read
ai safetybenchmarkingmulti-agent systems
A humanoid robot or animated figure demonstrating fluid motion gestures, with flowing digital data streams or code…
AI Research
75

LLaMo: The First Truly Unified Motion-Language AI Model That Understands and Generates Human Movement

Researchers have developed LLaMo, a groundbreaking AI model that unifies motion understanding and generation with language capabilities. Unlike previous approaches that suffered from catastrophic forgetting, LLaMo preserves linguistic knowledge while achieving real-time motion generation at over 30 FPS.

arxiv.org/Feb 12, 2026/3 min read
natural language processingcomputer visionhuman motion analysis