Knowledge catalogue
arXiv:2602.11675v3 Announce Type: replace Abstract: Large language models may answer causal questions correctly for the wrong reasons, substituting associational shortcuts P(Y|X) for the interventiona
arXiv:2604.18982v1 Announce Type: new Abstract: Social intelligence, the ability to navigate complex interpersonal interactions, presents a fundamental challenge for language agents. Training such age
arXiv:2604.19139v1 Announce Type: cross Abstract: As Large Language Models (LLMs) continue to evolve through alignment techniques such as Reinforcement Learning from Human Feedback (RLHF) and Constitu
arXiv:2604.19697v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have shown promising reasoning abilities, yet evaluating their performance in specialized domains remains chall
We need open traces so that everyone can train open agent models! cc @steipete @badlogicgames @thdxr @matanSF @hwchase17 People are misreading the SpaceX/Cursor deal as an M&A story. It’s actually a b
Succeeding in the agentic era requires a transformation in your data strategy: moving from human-scale to agent-first workloads, evolving from reactive intelligence to proactive action, and shifting f
Companies are shifting from gen AI that simply answers questions to autonomous agents that perceive, reason, and act on their behalf. Attempting to scale these agents on legacy stacks exposes structur
... which keeps things confusing, since it raises an important new question https://x.com/simonw/status/2046798283700617267 @TheAmolAvasare If I sign up for a new $20/month account today and roll the
Xiaomi's MiMo-V2.5 and MiMo-V2.5-Pro are both now available in Hermes Agent through Nous Portal and OpenRouter! Just `hermes update`! Xiaomi MiMo-V2.5 Series: Pushing Open-Source Agents Forward 🔸 MiMo
arXiv:2604.16729v1 Announce Type: new Abstract: State-of-the-art large language models (LLMs) show high performance in general visual question answering. However, a fundamental limitation remains: cur
arXiv:2604.16504v1 Announce Type: new Abstract: Manual digitisation of structured handwritten documents is slow and costly. We benchmark 17 leading frontier multi-modal large language models and open-
I find that open weights models over-perform on benchmarks compared to actual real-world usage, and Kimi feels like no exception. For example, a small amount of use will show that Kimi is not as good
Introducing ml-intern, the agent that just automated the post-training team @huggingface It's an open-source implementation of the real research loop that our ML researchers do every day. You give it
Love this work from Aksel and the post-training team at Hugging Face! Turns out the HF ecosystem (papers, datasets, models all accessible through CLI, skills and md files) is perfect for running SOTA
arXiv:2604.16506v1 Announce Type: cross Abstract: Large language models perform well on many medical QA benchmarks, but real clinical reasoning often requires integrating evidence across multiple imag
My hunch for now is that this was an ill-considered test which they didn't anticipate would be instantly spotted and cause (justified) uproar - here's hoping they decide that the 'test' isn't a good i
OpenAI Group PBC today launched ChatGPT Images 2.0, an upgraded version of the image generator built into its popular chatbot. The company also debuted a new technical training service called Codex La
arXiv:2604.16515v1 Announce Type: new Abstract: The rapid proliferation of Multimodal Large Language Models (MLLMs) has enabled mobile agents to execute high-stakes financial transactions, but their a
arXiv:2604.17715v1 Announce Type: cross Abstract: Recent advances in large language models for test case generation have improved branch coverage via prompt-engineered mutations. However, they still l
arXiv:2604.16529v1 Announce Type: cross Abstract: Test-time scaling has become a powerful way to improve large language models. However, existing methods are best suited to short, bounded outputs that
arXiv:2505.19897v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have extended their impact beyond Natural Language Processing, substantially fostering the development of interdi
arXiv:2604.17159v1 Announce Type: cross Abstract: We present, to our knowledge, the most comprehensive cross-model evaluation of LLM agents on offensive cybersecurity tasks, benchmarking 10 frontier m
arXiv:2604.18478v1 Announce Type: cross Abstract: Persistent memory is the bottleneck separating stateless chatbots from long-running agentic systems. Retrieval-augmented generation (RAG) over flat ve
arXiv:2503.14281v4 Announce Type: replace-cross Abstract: AI coding assistants are widely used for tasks like code generation. These tools now require large and complex contexts, automatically sourced
arXiv:2510.06708v3 Announce Type: replace-cross Abstract: Conducting systematic reviews is laborious. In the screening or study selection phase, the number of papers can be overwhelming. Recent resear
A sudden outage impacted both ChatGPT and OpenAI Codex services on April 20, 2026, creating problems for thousands of users globally. Conversations, logins, voice mode, image generation, Codex, and th
arXiv:2604.15336v1 Announce Type: cross Abstract: Large language models (LLMs) enable increasingly capable tutoring-style conversational agents, yet effective tutoring requires sensitivity to learners
arXiv:2406.15809v5 Announce Type: replace Abstract: Citizen reporting platforms help the public and authorities stay informed about sexual harassment incidents. However, the high volume of data shared
arXiv:2604.15384v1 Announce Type: cross Abstract: We introduce LinuxArena, a control setting in which agents operate directly on live, multi-service production environments. LinuxArena contains 20 env
arXiv:2604.15597v1 Announce Type: new Abstract: Large Language Models (LLMs) are poised to disrupt knowledge work, with the emergence of delegated work as a new interaction paradigm (e.g., vibe coding
arXiv:2604.16275v1 Announce Type: new Abstract: This paper explores the response of Large Language Models (LLMs) to user prompts with different degrees of politeness and impoliteness. The Politeness T
arXiv:2604.15937v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed to curate and rank human-created content, yet the nature and structure of their biases in these
Thanks for the shoutout re: ParseBench! 📑 🙂 Opus 4.7 is definitely a step up from Opus 4.6 on document understanding capabilities. Great to see labs start to push a bit more on this front. [16 Apr 202
The biggest challenge of using chat-based AI systems is that the details of what they can do are invisible - those tool descriptions are the missing manual, publishing them would be a huge benefit to
The continuing gap between the capabilities of Gemini Pro 3.1 (very good model) and the capabilities of the Gemini app/website is odd. The model can do what Claude/GPT can do, but there is a minimal h
arXiv:2511.15915v2 Announce Type: replace-cross Abstract: We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerator
arXiv:2603.23682v2 Announce Type: replace-cross Abstract: The rapid adoption of large language models (LLMs) in education raises profound challenges for assessment design. To adapt assessments to the
arXiv:2502.04689v4 Announce Type: replace Abstract: Intent, a critical cognitive notion and mental state, is ubiquitous in human communication and problem-solving. Accurately understanding the underly
This year's PyCon US is coming up next month from May 13th to May 19th, with the core conference talks from Friday 15th to Sunday 17th and tutorial and sprint days either side. It's in Long Beach, Cal
arXiv:2412.14751v2 Announce Type: replace Abstract: Retrieval-augmented generation (RAG) mitigates hallucination in Large Language Models (LLMs) by using query pipelines to retrieve relevant external
🇺🇸 xAI just dropped a real-time speech-to-text model built for voice apps: high limits, multi-language support, the works. Priced at just 0.10–0.20/hour, crushing most competitors. Quiet release, loud
Agent evals are drifting away from production reality. Most benchmarks use clean tasks, well-specified requirements, deterministic metrics, and retrospective curation. Production work is messier, with
arXiv:2604.13065v1 Announce Type: new Abstract: LLMs can execute every step of chain-of-thought reasoning correctly and still produce wrong final answers. We introduce the Novel Operator Test, a bench
Introduction Advances in AI model-powered exploitation have demonstrated that general-purpose AI models can excel at vulnerability discovery, even without being purpose-built for the task. Eventually,
arXiv:2602.24119v2 Announce Type: replace Abstract: Purpose: This study evaluates the quality of commercial large language model (LLM) machine translation (MT) for Ancient Greek technical prose and be
GLM-5.1 Tool Calling Issue Fix & Chat Template Update If you are running GLM-5.1 with vLLM/SGLang and using tool calling, please update your chat template. http://huggingface.co/zai-org/GLM-5.1/blob/m
arXiv:2604.13066v1 Announce Type: new Abstract: In-context learning has established itself as an important learning paradigm for Large Language Models (LLMs). In this paper, we demonstrate that LLMs c
New insane model from Jackrong on @huggingface 🤯 Qwen3.5-9B-GLM5.1-Distill-v1 🧠 Distilled on GLM-5.1 reasoning ⚙️ Deeper thinking than base model 🧪 Benchmarks coming soon ✅ Fits on 8GB VRAM ✍️ New mod
Response. Hey Ethan! Sean here, PM on http://Claude.ai - thanks for the feedback. This isn't a router, this is the model being trained to decide when to think based on the context -- we've been runnin
Shocking result on my pelican benchmark this morning, I got a better pelican from a 21GB local Qwen3.6-35B-A3B running on my laptop than I did from the new Opus 4.7! Qwen on the left, Opus on the righ
VLM Performance：Qwen3.6 is natively multimodal, and Qwen3.6-35B-A3B showcases perception and multimodal reasoning capabilities that far exceed what its size would suggest, with only around 3 billion a
We replicated Mythos findings in opencode using public models, not Anthropic's private stack. The moat is moving from model access to validation: finding vulnerability signal is getting cheaper; turni
arXiv:2512.20798v4 Announce Type: replace Abstract: As autonomous AI agents are deployed in high-stakes environments, ensuring their safety has become a paramount concern. Existing safety benchmarks p
arXiv:2604.12162v1 Announce Type: new Abstract: The rapid deployment of AI agents in commercial settings has outpaced the development of evaluation methodologies that reflect production realities. Exi
arXiv:2604.11950v1 Announce Type: cross Abstract: While recent LLM-based agents can identify many candidate bugs in source code, their reports remain static hypotheses that require manual validation,
btw the famous slack chart is slack propaganda and everyone who cites it is legally obligated to also link to @sophiebits Every time I see a tweet saying “I can vibe code this in a weekend” - I think
Budget blown on closed APIs is a solvable problem. Simply replace 20m of closed-source tokens with 1m on Minimax M2.7. Frontier performance. 20x lower cost. No rate limits. If you're rethinking your A
arXiv:2604.12743v1 Announce Type: new Abstract: While recent research has explored AI tools' ability to classify the quality of mathematical tasks (arXiv:2603.03512), little is known about their capac
arXiv:2603.24647v4 Announce Type: replace Abstract: The autoresearch repository enables an LLM agent to optimize hyperparameters by editing training code directly. We use it as a testbed to compare cl
arXiv:2604.12376v1 Announce Type: cross Abstract: When LLM conversations grow beyond the context window, old content must be evicted -- but how does the model recover it when needed? We propose cooper