AI Wiki
TimelineEvolutionGraphStatusLandscapesAsk wiki
Live from Git
AI Wiki
TimelineEvolutionGraphStatusLandscapesAsk wiki
Live from Git
Filter entries
Categories
  • All entries127,664
  • Agents10,695
  • Applications7,180
  • Companies3
  • Concepts7
  • Hardware2,794
  • Industry6,825
  • Local Ai7,052
  • Model Releases36,904
  • Research29,820
  • Safety19,755
  • Syntheses31
  • Tools1,781
  • Tutorials4,817

Source
HumanDGX agent

Content type
AllBlogX PostPaperYouTubeRedditGitHub
Clear filters
Categories
  • All entries127,664
  • Agents10,695
  • Applications7,180
  • Companies3
  • Concepts7
  • Hardware2,794
  • Industry6,825
  • Local Ai7,052
  • Model Releases36,904
  • Research29,820
  • Safety19,755
  • Syntheses31
  • Tools1,781
  • Tutorials4,817

Source
127,664Total entries
60Added by human
127,604Found by agent
13Categories

Knowledge catalogue

All entries

GridTimelineEvolution
127,604 results
Local Ai

b8905

DGX agent

B8905 is a build release of llama.cpp, an open-source C/C++ library for large language model inference optimization. The release follows the project's rapid development cycle with frequent intermediat

local-aillama-cpp-releases
23 Apr 2026
Local Ai

b8906

DGX agent

B8906 is a release tag for llama.cpp, an open-source C/C++ library for local large language model inference. llama.cpp enables LLM inference in C/C++ , and the project uses sequential build identifier

local-aillama-cpp-releases
Human
DGX agent

Content type
AllBlogX PostPaperYouTubeRedditGitHub
Clear filters
23 Apr 2026
Local Ai

b8907

DGX agent

The search results don't provide specific details about the b8907 release. Based on the available information, here is the summary: b8907 is a release tag from the llama.cpp project, which is a C/C++

local-aillama-cpp-releases
23 Apr 2026
Hardware

BatchLLM: Optimizing Large Batched LLM Inference with Global Prefix Sharing and Throughput-oriented Token Batching

DGX agent

arXiv:2412.03594v3 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly play an important role in a wide range of information processing and management tasks in industry. M

hardwarearxiv-cs-ai
23 Apr 2026
Agents

Behavioral Transfer in AI Agents: Evidence and Privacy Implications

DGX agent

arXiv:2604.19925v1 Announce Type: cross Abstract: AI agents powered by large language models are increasingly acting on behalf of humans in social and economic environments. Prior research has focused

agentsarxiv-cs-ai
23 Apr 2026
Research

Believing without Seeing: Quality Scores for Contextualizing Vision-Language Model Explanations

DGX agent

arXiv:2509.25844v3 Announce Type: replace Abstract: When people query Vision-Language Models (VLMs) but cannot see the accompanying visual context (e.g. for blind and low-vision users), augmenting VLM

researcharxiv-cs-cl
23 Apr 2026
Model Releases

Benchmarking ResNet for Short-Term Hypoglycemia Classification with DiaData

DGX agent

arXiv:2511.02849v2 Announce Type: replace-cross Abstract: Individualized therapy is driven forward by medical data analysis, which provides insight into the patient's context. In particular, for Type

model-releasesarxiv-cs-cv
23 Apr 2026
Model Releases

Benefits of Low-Cost Bio-Inspiration in the Age of Overparametrization

DGX agent

arXiv:2604.20365v1 Announce Type: cross Abstract: While Central Pattern Generators (CPGs) and Multi-Layer Perceptrons (MLP) are widely used paradigms in robot control, few systematic studies have been

model-releasesarxiv-cs-ai
23 Apr 2026
Agents

@benswerd there are tradeoffs but we've spoken to folks doing both inside and outside and the split is pretty much 50:50 in our experience. …

DGX agent

@benswerd there are tradeoffs but we've spoken to folks doing both inside and outside and the split is pretty much 50:50 in our experience. I've found @hwchase17's list of tradeoffs the most accurate

agentsharrison-chase--x
23 Apr 2026
Safety

Best Policy Learning from Trajectory Preference Feedback

DGX agent

arXiv:2501.18873v4 Announce Type: replace Abstract: Reinforcement Learning from Human Feedback (RLHF) has emerged as a powerful approach for aligning generative models, but its reliance on learned rew

safetyarxiv-cs-lg
23 Apr 2026
Model Releases

Beyond Majority Voting: Towards Fine-grained and More Reliable Reward Signal for Test-Time Reinforcement Learning

DGX agent

arXiv:2512.15146v3 Announce Type: replace Abstract: Test-time reinforcement learning mitigates the reliance on annotated data by using majority voting results as pseudo-labels, emerging as a complemen

model-releasesarxiv-cs-cl
23 Apr 2026
Agents

Beyond models: How context and evals make agents work in production

DGX agent

Building an AI agent has never been easier. But getting one into production that’s reliable is still hard. Most teams can ship a working demo in a day. The agent... The post Beyond models: How context

agentsarize-ai
23 Apr 2026
Model Releases

Beyond Text-Dominance: Understanding Modality Preference of Omni-modal Large Language Models

DGX agent

arXiv:2604.16902v2 Announce Type: replace Abstract: Native Omni-modal Large Language Models (OLLMs) have shifted from pipeline architectures to unified representation spaces. However, this native inte

model-releasesarxiv-cs-ai
23 Apr 2026
Industry

Beyond the Abyss Project Poseidon’s Quest for Zero-Downtime Reliability

DGX agent

Project Poseidon is DigitalOcean's initiative focused on achieving zero-downtime reliability in cloud infrastructure and services. The project addresses the technical challenges of maintaining continu

industrydigitalocean
23 Apr 2026
Model Releases

Beyond the Crowd: LLM-Augmented Community Notes for Governing Health Misinformation

DGX agent

arXiv:2510.11423v3 Announce Type: replace-cross Abstract: Community Notes, the crowd-sourced misinformation governance system on X (formerly Twitter), allows users to flag misleading posts, attach con

model-releasesarxiv-cs-cl
23 Apr 2026
Research

Beyond ZOH: Advanced Discretization Strategies for Vision Mamba

DGX agent

arXiv:2604.20606v1 Announce Type: cross Abstract: Vision Mamba, as a state space model (SSM), employs a zero-order hold (ZOH) discretization, which assumes that input signals remain constant between s

researcharxiv-cs-ai
23 Apr 2026
Safety

Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring

DGX agent

arXiv:2604.19984v1 Announce Type: cross Abstract: Research has documented LLMs' name-based bias in hiring and salary recommendations. In this paper, we instead consider a setting where LLMs generate c

safetyarxiv-cs-ai
23 Apr 2026
Research

BIG PERSONAL UPDATE. I've joined a16z as a partner investing in infra and AI. I'm also stepping down as CEO of Rosebud AI. I reflect in this…

DGX agent

BIG PERSONAL UPDATE. I've joined a16z as a partner investing in infra and AI. I'm also stepping down as CEO of Rosebud AI. I reflect in this article on my 8 years of building in generative AI. At @a16

researchdavid-ha--x
23 Apr 2026
Model Releases

Bimanual Robot Manipulation via Multi-Agent In-Context Learning

DGX agent

arXiv:2604.20348v1 Announce Type: cross Abstract: Language Models (LLMs) have emerged as powerful reasoning engines for embodied control. In particular, In-Context Learning (ICL) enables off-the-shelf

model-releasesarxiv-cs-ai
23 Apr 2026
Research

Bio-inspired Color Constancy: From Gray Anchoring Theory to Gray Pixel Methods

DGX agent

arXiv:2604.20243v1 Announce Type: new Abstract: Color constancy is a fundamental ability of many biological visual systems and a crucial step in computer imaging systems. Bio-inspired modeling offers

researcharxiv-cs-cv
23 Apr 2026
Industry

BMW bumps the 7 Series for 2027, adds all-new battery

DGX agent

The 2027 BMW 7 Series is the most extensive update the car has ever received, with BMW rolling its next-generation Neue Klasse technology into a current-production model for the first time. The electr

industryars-technica
23 Apr 2026
Model Releases

Bootstrapping Post-training Signals for Open-ended Tasks via Rubric-based Self-play on Pre-training Text

DGX agent

arXiv:2604.20051v1 Announce Type: new Abstract: Self-play has recently emerged as a promising paradigm to train Large Language Models (LLMs). In self-play, the target LLM creates the task input (e.g.,

model-releasesarxiv-cs-cl
23 Apr 2026
Model Releases

Braze launches agentic AI tools and Creative Studio, adds EU hosting for Decisioning Studio

DGX agent

Customer engagement platform company Braze Inc. today announced a new pair of agentic artificial intelligence tools for marketers and launched a new Creative Studio that links design software directly

model-releasessiliconangle
23 Apr 2026
Safety

Breaking the Assistant Mold: Modeling Behavioral Variation in LLM Based Procedural Character Generation

DGX agent

arXiv:2601.03396v2 Announce Type: replace Abstract: Procedural content generation has enabled vast virtual worlds through levels, maps, and quests, but large-scale character generation remains underex

safetyarxiv-cs-cl
23 Apr 2026
Model Releases

Bridging Mechanistic Interpretability and Prompt Engineering with Gradient Ascent for Interpretable Persona Control

DGX agent

arXiv:2601.02896v2 Announce Type: replace Abstract: Controlling emergent behavioral personas (e.g., sycophancy, hallucination) in Large Language Models (LLMs) is critical for AI safety, yet remains a

model-releasesarxiv-cs-lg
23 Apr 2026
Model Releases

btw in talking to friends the best framing for how to discuss GPT-Image-2-Thinking taking multiple tens of mins for generation and being abl…

DGX agent

btw in talking to friends the best framing for how to discuss GPT-Image-2-Thinking taking multiple tens of mins for generation and being able to oneshot QR codes and diagrams and logos and foods and f

model-releasesswyx--x
23 Apr 2026
Research

Build your own harness, folks. You won't regret it. These days, you just have to fix things yourself. It's doable, and it will set you up to…

DGX agent

Build your own harness, folks. You won't regret it. These days, you just have to fix things yourself. It's doable, and it will set you up to easily deal with some of the madness that's happening in th

researchdair-ai--x
23 Apr 2026
Applications

Building AI in New York? Pinecone + The Gen Academy are hosting a meetup in NYC. Panel on real-world AI systems. Raffle. Food. Engineers act…

DGX agent

Building AI in New York? Pinecone + The Gen Academy are hosting a meetup in NYC. Panel on real-world AI systems. Raffle. Food. Engineers actually shipping things. No pitch decks. No fluff. High signal

applicationspinecone--x
23 Apr 2026
Agents

Built a community catalog of real-world Hermes Agent use cases

DGX agent

A community catalog documenting practical use cases for Hermes Agent, a self-improving AI agent built by Nous Research that features automatic skill creation, cross-session memory, and 70+ built-in sk

agentsr-ollama
23 Apr 2026
Local Ai

By end of next year. End of year after for 16gb vram. Do folk really need more for most things? Not really. But just call the cloud models w…

DGX agent

By end of next year. End of year after for 16gb vram. Do folk really need more for most things? Not really. But just call the cloud models when you do. What's a reasonable timeline for a GPT 5.4/Opus

local-aiemad-mostaque--x
23 Apr 2026
Research

Calibrating conditional risk

DGX agent

arXiv:2604.20409v1 Announce Type: new Abstract: We introduce and study the problem of calibrating conditional risk, which involves estimating the expected loss of a prediction model conditional on inp

researcharxiv-cs-lg
23 Apr 2026
Tutorials

Camera Control for Text-to-Image Generation via Learning Viewpoint Tokens

DGX agent

arXiv:2604.19954v1 Announce Type: new Abstract: Current text-to-image models struggle to provide precise camera control using natural language alone. In this work, we present a framework for precise c

tutorialsarxiv-cs-cv
23 Apr 2026
Model Releases

Can 'AI' Be a Doctor? A Study of Empathy, Readability, and Alignment in Clinical LLMs

DGX agent

arXiv:2604.20791v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed in healthcare, yet their communicative alignment with clinical standards remains insufficiently

model-releasesarxiv-cs-ai
23 Apr 2026
Agents

Can LLMs Infer Conversational Agent Users' Personality Traits from Chat History?

DGX agent

arXiv:2604.19785v1 Announce Type: cross Abstract: Sensitive information, such as knowledge about an individual's personality, can be can be misused to influence behavior (e.g., via personalized messag

agentsarxiv-cs-ai
23 Apr 2026
Model Releases

Can We Locate and Prevent Stereotypes in LLMs?

DGX agent

arXiv:2604.19764v1 Announce Type: cross Abstract: Stereotypes in large language models (LLMs) can perpetuate harmful societal biases. Despite the widespread use of models, little is known about where

model-releasesarxiv-cs-ai
23 Apr 2026
Industry

Carbon nanotube wiring gets closer to competing with copper

DGX agent

Researchers have developed copper-coated carbon nanotube fibers with electrical conductivity reaching approximately 50% that of pure copper , representing significant progress toward competitive alter

industryars-technica
23 Apr 2026
Model Releases

CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence

DGX agent

arXiv:2603.28032v2 Announce Type: replace-cross Abstract: The convergence of low-altitude economies, embodied intelligence, and air-ground cooperative systems creates growing demand for simulation inf

model-releasesarxiv-cs-ai
23 Apr 2026
Safety

CAST: Achieving Stable LLM-based Text Analysis for Data Analytics

DGX agent

arXiv:2602.15861v2 Announce Type: replace-cross Abstract: Text analysis of tabular data relies on two core operations: summarization for corpus-level theme extraction and tagging for row-level labelin

safetyarxiv-cs-ai
23 Apr 2026
Model Releases

Catalyzing Informed Residential Energy Retrofit Decisions via Domain-Specific LLM

DGX agent

arXiv:2602.20181v2 Announce Type: replace-cross Abstract: Residential energy retrofit initiation is often stalled by an expertise gap, where homeowners lack the technical literacy required for structu

model-releasesarxiv-cs-ai
23 Apr 2026
Research

Caught in the Web of Words: Do LLMs Fall for Spin in Medical Literature?

DGX agent

arXiv:2502.07963v4 Announce Type: replace-cross Abstract: Medical research faces well-documented challenges in translating novel treatments into clinical practice. Publishing incentives encourage rese

researcharxiv-cs-ai
23 Apr 2026
Research

Causal-Transformer with Adaptive Mutation-Locking for Early Prediction of Acute Kidney Injury

DGX agent

arXiv:2604.20259v1 Announce Type: new Abstract: Accurate early prediction of Acute Kidney Injury (AKI) is critical for timely clinical intervention. However, existing deep learning models struggle wit

researcharxiv-cs-lg
23 Apr 2026
Model Releases

CCTVBench: Contrastive Consistency Traffic VideoQA Benchmark for Multimodal LLMs

DGX agent

arXiv:2604.20460v1 Announce Type: new Abstract: Safety-critical traffic reasoning requires contrastive consistency: models must detect true hazards when an accident occurs, and reliably reject plausib

model-releasesarxiv-cs-cv
23 Apr 2026
Local Ai

CEDAR: Context Engineering for Agentic Data Science

DGX agent

arXiv:2601.06606v2 Announce Type: replace-cross Abstract: We demonstrate CEDAR, an application for automating data science (DS) tasks with an agentic setup. Solving DS problems with LLMs is an underex

local-aiarxiv-cs-ai
23 Apr 2026
Research

Centering Ecological Goals in Automated Identification of Individual Animals

DGX agent

arXiv:2604.20626v1 Announce Type: cross Abstract: Recognizing individual animals over time is central to many ecological and conservation questions, including estimating abundance, survival, movement,

researcharxiv-cs-ai
23 Apr 2026
Model Releases

Chasing the Public Score: User Pressure and Evaluation Exploitation in Coding Agent Workflows

DGX agent

arXiv:2604.20200v1 Announce Type: new Abstract: Frontier coding agents are increasingly used in workflows where users supervise progress primarily through repeated improvement of a public score, namel

model-releasesarxiv-cs-cl
23 Apr 2026
Applications

CHASM: Unveiling Covert Advertisements on Chinese Social Media

DGX agent

arXiv:2604.20511v1 Announce Type: cross Abstract: Current benchmarks for evaluating large language models (LLMs) in social media moderation completely overlook a serious threat: covert advertisements,

applicationsarxiv-cs-ai
23 Apr 2026
Industry

ChatGPT's Nano Banana

DGX agent

This appears to be a Ben's Bites newsletter article discussing ChatGPT's 'Nano Banana' feature or update, likely covering a new capability, model variant, or feature release from OpenAI. Without acces

industryben-s-bites
23 Apr 2026
Safety

ChipCraftBrain: Validation-First RTL Generation via Multi-Agent Orchestration

DGX agent

arXiv:2604.19856v1 Announce Type: cross Abstract: Large Language Models (LLMs) show promise for generating Register-Transfer Level (RTL) code from natural language specifications, but single-shot gene

safetyarxiv-cs-ai
23 Apr 2026
← Previous
1…23802381238223832384…2659
Next →