Knowledge catalogue
arXiv:2609.04098v1 Announce Type: new Abstract: Hybrid LLMs pair softmax attention with linear-attention layers such as Gated DeltaNet (GDN), whose recurrent state summarizes the context in fixed size
arXiv:2606.21144v2 Announce Type: replace-cross Abstract: Long-term memory systems allow LLM agents to preserve information beyond a single context window, but most systems focus on storing and retrie
arXiv:2609.02459v1 Announce Type: new Abstract: We present CivBench, an open-source benchmark for evaluating language model agents in long-horizon, tool-mediated environments through the Model Context
arXiv:2603.19539v2 Announce Type: replace-cross Abstract: We introduce an expert curated, real-world benchmark for evaluating document-grounded question-answering (QA) motivated by generic drug assess
AI models have clearly proven their ability to discover and exploit vulnerabilities without much, if any, human assistance. To help defenders gain the advantage with AI, we built the Mantis harness to
Hi, I've noticed that the model often sees 'garbled text' in its context. Sometimes it declare that the tools instructions are corrupted, sometimes it is the content of some .md files, ora other files
qwen4exp: fix seq_cp, block position keying, mtmd input, cuda abort, add tests (#27941) qwen4exp: follow up fixes -kvu NaN collapse fix Assisted-by: Claude indexer cache ext.x/ext.y restore fix Assist
arXiv:2608.30653v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) are increasingly deployed in multi-image scenarios requiring complex reasoning across visual contexts. Howeve
arXiv:2608.29884v1 Announce Type: new Abstract: We show that sequence-level distillation from a capable long-context teacher model is a simple, annotation-free, and data-efficient strategy for improvi
arXiv:2608.29798v1 Announce Type: cross Abstract: The same Persona behavior can be beneficial in one context but harmful in another, causing static Persona elicitation to perform inconsistently across
arXiv:2608.30310v1 Announce Type: cross Abstract: Hybrid large language models interleave full-attention layers with linear-attention layers to reduce the cost of long-context inference. This structur
arXiv:2608.30567v1 Announce Type: new Abstract: We present Turing-20B-A2B, a 20B-parameter Mixture-of-Experts language model that activates approximately 2B parameters per token, designed for long-con
arXiv:2604.02194v2 Announce Type: replace-cross Abstract: Retrieval-Augmented Language Models (RALMs) have shown strong potential in knowledge-intensive tasks, yet they remain vulnerable when retrieve
arXiv:2502.09245v3 Announce Type: replace-cross Abstract: In contrast to RNNs, which compress their history into a single hidden state, Transformers can attend to all past tokens directly. However, st
arXiv:2608.28044v1 Announce Type: cross Abstract: Large language model (LLM) inference serving is priced by tokens, but GPU energy is consumed over inference windows. This accounting mismatch makes to
arXiv:2608.28460v1 Announce Type: new Abstract: Autoregressive video diffusion enables scalable long-video generation by producing chunks from a bounded recent context. While recency-based caching pre
My conversation with @jerryjliu0. He's building @llama_index: document infrastructure for AI agents, the layer underneath everything models need to work. Did you know he's the guy whose tool half the
I'm running Muse Glimmer 30B EXL3-SC 3.00bpw H4, fully resident on my 12GB VRAM GPU at 100K context with Q8_O KV cache. It's a joy to use a dense 30B model at this size and still get 30 tok/s on a VRA
arXiv:2608.26172v1 Announce Type: cross Abstract: Deploying natural-language interfaces over enterprise OLTP catalogs fails at scale because semantic parsers collapse under schema-graph scaling, infla
arXiv:2608.26175v1 Announce Type: cross Abstract: Extractive prompt compression promises to cut LLM inference costs by removing low-information tokens, and learned compressors such as LLMLingua-2 repo
arXiv:2608.25570v1 Announce Type: new Abstract: Hardware kernel optimization requires repeated compilation, correctness testing, profiling, and revision. LLM agents can automate parts of this process,
arXiv:2608.25531v1 Announce Type: new Abstract: Humanities and social science research requires close reading of long narrative materials such as novels, scripts, archives, and case reports, yet many
arXiv:2608.24905v1 Announce Type: cross Abstract: Service robots may encounter ambiguous user requests that require context-aware inference. Users may also have unique preferences with certain tasks w
arXiv:2608.25236v1 Announce Type: cross Abstract: Clinical decision-making often involves prioritizing ethical values, such as beneficence, non-maleficence, respecting a patient's autonomy, and justic
arXiv:2608.17889v1 Announce Type: cross Abstract: Visually rich documents encode relevance through language, layout, structured visual elements, and corpus context, yet retrieval is typically evaluate
https://preview.redd.it/i8rjx0ar5mlh1.png?width=2160&format=png&auto=webp&s=c2588bc7b2519ea71b176ca73faf566dfc585496 I wanted to see whether a heavily quantized 27B model running entirely on an older
arXiv:2608.23503v1 Announce Type: new Abstract: Text-based person anomaly search requires distinguishing individuals based on fine-grained, context-dependent behaviors rather than mere appearance. Exi
arXiv:2608.22452v1 Announce Type: new Abstract: Surprisal, the negative log-probability a language model assigns to a word given its preceding context, reliably predicts adult reading times. Does it c
If you keep safety rules or coding standards in an AGENTS.md or a CLAUDE.md, this one is worth your time. (bookmark it) Researchers measured what context compaction actually destroys across 20 product
arXiv:2608.23552v1 Announce Type: new Abstract: Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context
In this installment of Rise of the AI Engineer, Stuart Sy from OpenAI, explains why the bottleneck has moved off the model and onto context, evals, and observability. The post Why better models don’t
arXiv:2608.20481v1 Announce Type: cross Abstract: The Model Context Protocol (MCP) is an open source JSON-RPC protocol that standardizes how large language models (LLMs) interact with external systems
arXiv:2608.20820v1 Announce Type: new Abstract: Large language models (LLMs) are vulnerable to multi-turn jailbreak attacks that progressively manipulate conversation context. Existing certified robus
arXiv:2608.21265v1 Announce Type: new Abstract: Large language models often rely on Chain-of-Thought (CoT) reasoning to solve complex tasks, but verbose reasoning traces introduce substantial inferenc
arXiv:2608.21308v1 Announce Type: new Abstract: Test-Time Training (TTT) enables long-context processing via continuous weight updates during inference, but current methods struggle to balance the exp
arXiv:2608.20341v1 Announce Type: new Abstract: Frontier coding agents backed by large language models with context windows from hundreds of thousands to millions of tokens are restructuring the Softw
What a fascinating paper on AI agents. A lot of the issues we see with AI agents today revolve around wrong assumptions the LLMs make. This leads to problems like hallucination, cost inefficiencies, u
arXiv:2608.19395v1 Announce Type: cross Abstract: Hybrid Transformer-Mamba large language models (LLMs) enhance long-context efficiency, but their heterogeneous computation and communication patterns
arXiv:2608.19993v1 Announce Type: new Abstract: Loading reusable skill documents into a bounded context window is now the primary way large language model (LLM) agents acquire task-specific capabiliti
Hi fellows fully-local halos, after manually following existing guides, I decided to build an LLM API endpoint installation and optimization guide that works even when autonomously followed by my pi a
Not everyone has the disposable income to build a small data center, so making this post for the underdogs as I was very surprised by the performance/results of this 35B MOE model. Context is admitted
arXiv:2608.18667v1 Announce Type: new Abstract: Modern deep learning architectures have demonstrated strong performance in dental CBCT segmentation. One remaining crucial challenge is accurate tooth l
arXiv:2608.18144v1 Announce Type: cross Abstract: Modal auxiliaries such as must, should, and have to mark necessity and obligation within the contexts of speaker authority and interpersonal stance. W
Adronite Inc. today launched Codistry, an artificial intelligence coding platform for large enterprise codebases. The platform runs on Adronite’s Context Engine, or ACE, which the company has patented
arXiv:2608.17528v1 Announce Type: new Abstract: Modern agents operate inside agent harnesses that manage tools, context, and control flow, making the harness a critical part of the agent system. Our o
arXiv:2608.17896v1 Announce Type: new Abstract: Recurrent models process long contexts efficiently by compressing their history into a fixed-size state, but modern architectures typically do so in a s
arXiv:2608.16178v1 Announce Type: cross Abstract: Operational telemetry is predominantly engineered for human reading: systems repeatedly serialize verbose prose, static keys, and redundant context ac
arXiv:2608.14950v1 Announce Type: new Abstract: Generative AI systems are increasingly producing real-world artifacts, however their efficacy and validity are often evaluated via context-free LLM-scor
Written by: Alex Tselevich, Michael Maturi Introduction Adversarial misuse of AI has increased the risk of data theft and extortion events, because when proprietary source code is exposed, defenders m
arXiv:2608.13568v1 Announce Type: cross Abstract: Coding agents spend most of their context budget on retrieval. Lexical retrieval (grep) is universal, instant, and zero-setup, but noisy: it cannot te
arXiv:2608.13668v1 Announce Type: new Abstract: Linear attention keeps a state of fixed size. At long context, many stored items share this state, and interference between them degrades retrieval. Gat
Friday's big release was Qwen 3.8 27B, an Apache 2 licensed 27B parameter vision-capable LLM from Alibaba's Qwen research lab. I've been looking forward to this one: 27B is an excellent size for runni
gave a talk 'owning your intelligence' - ty @sequoia @sonyatweetybird for having me talked about harnesses and evals and the role they play in owning your intelligence TLDR: > agents = model + harness
arXiv:2608.01735v2 Announce Type: replace Abstract: On-policy (self) distillation (OPSD) is increasingly adopted for language-model post-training. It strengthens the teacher with privileged informatio
arXiv:2608.10256v1 Announce Type: new Abstract: Accurate vessel trajectory prediction is critical for maritime safety and anomaly detection, yet existing models often struggle with geographic bias and
Alibaba released the open‑weights Qwen3.8‑2.4T‑A95B (Qwen3.8‑Max), a fine‑grained mixture‑of‑experts model with 2.4 trillion parameters, hybrid full‑ and linear‑attention, a one‑million‑token context
Public benchmarks can show how a model performs in general. Production reliability depends on the context and harness around it, which only your team can evaluate against its own data, workflows, and
arXiv:2608.08575v1 Announce Type: cross Abstract: Accurate 3D medical image segmentation requires the integration of long-range anatomical context with fine boundary detail. Existing methods often mod
arXiv:2608.08939v1 Announce Type: new Abstract: The rise of autonomous AI agents represents a major paradigm shift in how users interact with mobile devices. Frameworks such as MobileRun and Mobile-Us
arXiv:2608.08389v1 Announce Type: new Abstract: Long-horizon research agents solve open-ended tasks through iterative retrieval, aggregation, and synthesis, but context grows rapidly while the margina