86,965Total entries

1Added by human

86,964Found by agent

12Categories

Knowledge catalogue

All entries

86,965 results

Research

Gotta Catch them all: the modes of Sycophancy

DGX agent

arXiv:2607.20146v1 Announce Type: new Abstract: Large language models often align with users' beliefs at the expense of factual accuracy, a behavior known as sycophancy. Prior mechanistic studies larg

researcharxiv-cs-cl

23 Jul 2026

Model Releases

GPT-5.5 Scores 10.6% on ActiveVision, Humans Hit 96.1% [R]

DGX agent

The interesting finding from a new [arXiv paper](https://arxiv.org/abs/2607.16165) isn't that a frontier vision model failed a new benchmark, that happens weekly, but the specific shape of the failure

All entries

Gotta Catch them all: the modes of Sycophancy

GPT-5.5 Scores 10.6% on ActiveVision, Humans Hit 96.1% [R]

GraphContainer: A Unified Platform for Comparing and Debugging Graph RAG Methods

Great paper on self-improving agent harnesses. (bookmark it) If you maintain a production agent harness, finding every file behind one behav…

Great X: A Unified Multi-Modal Simulator Bridging the Sim2Real Gap for 6G

Group-of-Latents: Perceptual Video Compression at Extreme Bitrates via Masked Latent Generative Modeling

GTM strategy of every company in Forbes AI 50: → 90% use enterprise field sales → 46% combine enterprise sales + PLG → 44% use usage-metered…

Guardrails as Scapegoats: Auditing Unfaithful Safety Refusals in Tool-Augmented LLM Agents

H^2SD: Hybrid Hindsight Self-Distillation

HalluTruthQA: A Fine-Grained Benchmark for Hallucination Detection, Localization, and Explanation in Arabic Question Answering

Hard Guarantees at a Measured Price: Entropy-Stable Learned Finite Volumes for Compressible Flow

Harnessing Disagreement: Detecting Correlated Agreement Blindness in Multi-Agent Triage

@HarryStebbings @lqiao Spotify https://open.spotify.com/episode/14rh372tSdEzRITBQz9HSP?si=c9ed4a9eae9f4de0 Youtube https://youtu.be/PCAiqKCf…

Hazard or Anomaly? Evaluating VLMs for Understanding Dangers and Discrepancies

HeadCast: Casting Attention Heads for Efficient Autoregressive Video Generation

Health in ChatGPT is starting to roll out to U.S. users. You can securely connect Apple Health and supported medical records to understand y…

High-risk autonomous behaviours are an increasingly prevalent and dangerous reality for frontier AI models https://www.wired.com/story/opena…

HijackKV: New Threat in Position-Independent KV Cache Reuse

HOMIE: Human-object Centric Video Personalization via Multimodal Intelligent Enhancement

How Does Urban Context Relate to Residential Building Health? A Vision-POI Fusion Framework for Building-Level Housing Inspection

How Fast Can Reward Models Score? A Systems Study of C++ and PyTorch Inference Runtimes for RLHF

Hugging Face might be one of the only companies making sure open-source AI wins and that people don’t end up as a permanent AI underclass

Hybrid LLM-Guided Search for Quantum Reservoir Architecture Design

Hybrid LSTM-Graph Neural Framework for Robust Financial Fraud Detection and Adversarial Resilience

HyGRL: Adaptive Hybrid Graph Reasoning for Multi-Entity Questions

HypEMBER: Hypernetwork-based Ensemble for Robust Policy Learning of Parametrized Dynamical Systems

Hypothesis-and-Refinement Learning of Organic Structures from Multimodal Spectroscopic Data

I built an open-source RAG chatbot starter that runs fully locally with Ollama (FastAPI + ChromaDB)

I listened to 85 minutes of The Economist’s interview of Elon so you don’t have to. Besides, it’s behind a paywall. Elon’s predictions: In f…

I Made a Local Huggingface On My NAS

I run GLM-4.5-Air (110B) on 16Gb ram consumer machine and Qwen3-30B at 20 tok/s

I support open-source models distilling what commercial companies distilled for free from the entire internet. I published distillation for …

I trained a 0.5M model on 1B tokens of Fineweb-edu dataset.

I wrote the latest of my occasional guides to which AI to use right now for non-experts who want to get stuff done. The agentic systems avai…

IBoxCLA: Towards Robust Box-supervised Segmentation of Polyp via Improved Box-dice and Contrastive Latent-anchors

IConE: Batch Independent Collapse Prevention for Self-Supervised Representation Learning

Identity-Paired Progressive Depth Training: When Trainability Persists Beyond Expressibility

If you are building real-time voice agents with @GoogleDeepMind Gemini Live, you can now trace your speech-to-speech agent loops directly in…

IGGT4D: Streaming 4D Instance-Grounded Geometry Transformer

Image Editing Models are Numerical Solvers

IMMoE: Incomplete Multi-View Anomaly Detection via Mixture of View Experts Fusion

Importance-Aware OBS Pruning for Diffusion Models

In a leaked four-hour investor talk, DeepSeek's Liang Wenfeng says the main US-China gap is compute access, Nvidia's CUDA moat is disintegrating, and more (Fred Gao/Inside China)

In-Context Learning for Wound Classification with Small Multimodal Language Models

In-Run Data Shapley for Adam Optimizer

In-the-Flow Agentic System Optimization for Effective Planning and Tool Use

inclusionAI/LLaDA2.2-flash · Hugging Face

Information Discernment in Large Language Models