Knowledge catalogue
Critical context on the new Anthropic blog: 1, AGI is *harder* than RSI (as used below). AGI: machine can do anything human can do, autonomously [not achieved] RSI (as used below): AI is a useful codi
Excited to share that I’ve joined OpenAI in London to work on pretraining! I’ve spent the last few years on pretraining and long-context, and I’ll be helping grow the London team. We’re hiring excepti
arXiv:2606.06087v1 Announce Type: new Abstract: Agent systems increasingly use textual skills to encode reusable task procedures, but injecting these skills into the prompt at every step incurs substa
5/5 Takeaway: pipeline order is a hyperparameter. If you're already paying for parallel rollouts, reuse them - they're relevant context, not just candidate answers. Full write-up: [https://www.ai21.co
arXiv:2603.09803v2 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in large language models but treats all correct solutions equally, potentia
We’ve been researching new ways for ChatGPT memory to carry context across conversations and keep it useful over time. Today, that work is rolling out as a more capable memory system in ChatGPT. https
Carl Franzen / VentureBeat: Alibaba releases Qwen3.7-Plus, a multimodal proprietary model with a 1M-token context window, costing $2 per 1M tokens, 60% less than text-only Qwen3.7-Max — However, like
arXiv:2606.03435v1 Announce Type: new Abstract: Cell Painting combines multiplexed fluorescent staining, high-content imaging, and quantitative analysis to generate high-dimensional phenotypic readout
Anthropic's Claude 3.5 Sonnet (M3) model features sparse attention mechanisms, 1 million token context length, and multimodal capabilities, with Together AI optimizing the serving infrastructure to en
Managed Deep Agents combine open-source Deep Agents with LangSmith's infrastructure to provide both the flexibility of open-source tools and the reliability of hosted execution with persistent context
Jeremy Howard announced the release of Vui, an open-source voice mode text-to-speech (TTS) model with 300 million parameters that can run on consumer GPUs and Apple Silicon. The model features context
arXiv:2510.00615v3 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed as agents in dynamic real-world environments, where success depends on maintaining precise re
Amazing deep dive from the @togethercompute team on serving MiniMax M3 in production. M3 with its 1M context, native multimodality and MiniMax Sparse Attention requires real work across paged decode,
arXiv:2606.01495v1 Announce Type: cross Abstract: We present CART (Context-Anchored Recurrent Transformer), a parameter-efficient language model that reuses a single shared core block R times across d
arXiv:2606.00872v1 Announce Type: new Abstract: AI-generated image detection is a moving-target problem: detectors trained on one generator often fail when a new generator appears, and only a few labe
MiniMax-M3 combines 1M context, native multimodality, and MiniMax Sparse Attention. The next layer is serving it efficiently: KV-block-major sparse attention, paged MSA decode, optimized index scoring
arXiv:2606.01923v1 Announce Type: new Abstract: Large Language Models (LLMs) frequently exhibit 'contextual disregard' when faced with input evidence that conflicts with their internal parametric memo
arXiv:2601.20803v2 Announce Type: replace Abstract: This paper presents several strategies to automatically obtain additional examples for in-context learning, effectively transforming relation extrac
arXiv:2606.02304v1 Announce Type: new Abstract: LLM-based agents can solve multi-step interactive tasks by combining reasoning with environment feedback, yet each episode starts from the same fixed co
arXiv:2606.00724v1 Announce Type: cross Abstract: Diffusion Large Language Models (DLMs) have demonstrated significant advantages across various tasks. However, constrained by their multi-step iterati
arXiv:2605.31272v1 Announce Type: new Abstract: As predictive models are increasingly deployed in high-stakes settings such as credit approval, there is a growing need for post-hoc methods that provid
arXiv:2605.31138v1 Announce Type: cross Abstract: User Experience Research (UXR) in a legal and regulatory contexts presents unique challenges that require specialised approaches to protect vulnerable
We'll get into @MiniMax_AI M3's model performance, the MSA architecture and what it means for long context, and how Together is optimizing inference and KV-cache for this new architecture. Set your re
We're going LIVE tomorrow with @togethercompute 🔥. @zpysky1125 is pulling back the curtain on M3: sparse attention, 1M context, all of it. You don't want to miss this. MiniMax M3 is live and Together
arXiv:2605.29971v1 Announce Type: new Abstract: Causal interventions in language model representations have largely targeted discrete features, like grammatical number. However, language models must a
arXiv:2605.30136v1 Announce Type: new Abstract: LLM-based multi-agent systems have demonstrated remarkable performance on complex tasks through collaborative reasoning. However, these systems tend to
arXiv:2601.03729v2 Announce Type: replace Abstract: Fine-grained recognition of marine organisms is important for ecological research, biodiversity monitoring, habitat conservation, and evidence-based
arXiv:2605.30283v1 Announce Type: new Abstract: MCP Server Proto-OKN (mcp-proto-okn) is a Python-based Model Context Protocol server that enables AI assistants to discover, inspect, query and integrat
arXiv:2605.30251v1 Announce Type: cross Abstract: Large language models (LLMs) often solve a task when all instructions are given in a single prompt, but fail when the same information is revealed gra
arXiv:2605.29863v1 Announce Type: new Abstract: Predicting the next mobile application a user will launch is essential for intelligent device resource management and proactive assistance. Existing mod
Important context for latest OpenAI announcement. Especially (5:30): 'The model spit out a long transcript of an answer. Then a team of expert mathematicians poured over this [transcript] and identifi
arXiv:2605.27389v1 Announce Type: cross Abstract: We study how conditioning context shapes personalization behavior in a teacher-facing educational recommender system. We compare contextual conditioni
Deep Agents v0.6 introduces ContextHubBackend, a versioned storage system for agent behavior files integrated with LangSmith Context Hub. This feature enables agents to access and manage the files and
arXiv:2605.28069v1 Announce Type: new Abstract: Adaptive context compression is vital for scaling Large Language Models (LLMs) to complex, multi-turn agent tasks. However, rule-based compression metho
arXiv:2605.26403v1 Announce Type: new Abstract: A long-standing goal of the research community is to develop highly interactive LLM-based dialogue agents. Recent research focuses on optimizing policie
Managed Deep Agents is built for agents that need to work over long time horizons, use tools, preserve context, and produce artifacts. A few examples of what teams are building: ✅ Support + triage age
arXiv:2601.18904v2 Announce Type: replace-cross Abstract: Auditory Large Language Models (LLMs) have demonstrated strong performance across a wide range of speech and audio understanding tasks. Nevert
arXiv:2604.18103v2 Announce Type: replace Abstract: Prefilling computational costs pose a significant bottleneck for Large Language Models (LLMs) and Large Multimodal Models (LMMs) in long-context set
arXiv:2605.26350v1 Announce Type: cross Abstract: In-context learning (ICL) is often motivated by the intuition that demonstrations help because they provide correct input-output examples. However, we
arXiv:2502.17666v4 Announce Type: replace-cross Abstract: Existing offline in-context reinforcement learning (ICRL) methods have predominantly relied on supervised training objectives, which are known
arXiv:2605.24248v1 Announce Type: cross Abstract: The Model Context Protocol (MCP) standardizes how a large-language-model (LLM) agent and an external tool server exchange messages, but not trust: a h
arXiv:2605.24267v1 Announce Type: new Abstract: Human conversation relies heavily on conversational implicature, in which speakers convey meanings that are suggested rather than explicitly stated. Alt
Infinite context windows seem to present a very large problem to using AI. Today's models already leak too much old information into current responses, a distraction that is part of why they are cogni
Alibaba's Qwen team has released Qwen3.7 Max, their most advanced model to date, now available for use with Go programming language support. The model features a 1 million token context window and is
your bank knows what you did. it has no idea why. that context, the why behind every transaction, is the most valuable data in banking. and nobody's capturing it. this changes with agentic banking. Me
arXiv:2605.23493v1 Announce Type: new Abstract: On-Policy Distillation (OPD) has gained wide attraction as an LLM post-training paradigm due to its effectiveness in improving capabilities without intr
arXiv:2603.19812v2 Announce Type: replace Abstract: To address this gap, we conduct a Virtual Reality experiment in which pedestrians interact with automated shuttles under varying approach angles (45
arXiv:2510.04567v2 Announce Type: replace-cross Abstract: Graph Neural Networks (GNNs) are powerful tools for processing relational data but often struggle to generalize to unseen graphs, giving rise
the @xai team really cooked with Grok Build the most beautiful (and performant) TUI I've seen so far >realtime context view >feels snappy in ghostty >first class worktree support >great resume session
arXiv:2605.22335v1 Announce Type: new Abstract: In-context learning for tabular data sets strong predictive standards in observational settings; it however primarily relies on correlational structure,
Essential context on OpenAI’s Erdos result I truly miss the age of science and transparency in AI. Especially given how much money and political power and governance and scientific understanding is at
Introducing Qwen3.7-Max from @Alibaba_Qwen, Qwen’s flagship model for the agent era with 1M context and leading performance across agentic coding, reasoning, and long-horizon autonomy. AI natives can
arXiv:2605.20994v1 Announce Type: new Abstract: Preference-based post-training aligns LLMs with human intent, yet safety behavior often remains brittle. A model may refuse a harmful request in a stand
as agents run for longer and with more context, there's a lot of tricky things you have to consider for deployment! super excited to dive into deploying Deep Agents for this Boston tech week meetup! .
Context compression isn't new in RAG. Our contribution is making it query-aware, citation-preserving, and fast enough for orchestration. Read the full research blog: https://research.perplexity.ai/art
This entry discusses foundational concepts and context related to the Culture series, particularly the concept of the General Systems Vehicle (GSV) from Iain M. Banks' science fiction novels. The piec
Healthspan just launched BioAge+, a biological age score that breaks down which systems are driving it and why kind of like a context graph for your health Introducing BioAge+. The first biological ag
arXiv:2605.18830v1 Announce Type: new Abstract: Regression and Bayesian accounts of in-context learning (ICL) explain how demonstrations can induce predictors, while mechanistic analyses often identif
arXiv:2605.17470v1 Announce Type: new Abstract: Image super-resolution (SR) aims to reconstruct high-quality, high-resolution (HR) images from low-resolution (LR) inputs and plays a critical role in v
For context, this is what an IMAX camera and an IMAX projector look like. I love how, at some point, we began to see all magnificent machines as such natural extensions of ourselves that they feel hum