Knowledge catalogue
arXiv:2607.21273v1 Announce Type: new Abstract: Dense per-step supervision is an appealing remedy for sparse-reward, long-horizon LLM agents: reward the agent for predicting its next observation, and
arXiv:2607.11149v3 Announce Type: replace Abstract: LLM agent benchmarks measure task completion, reliability, and inference cost, but not the persistent data an agent run leaves on disk, including lo
arXiv:2607.19366v1 Announce Type: new Abstract: Safety as Polytope (SaP) learns linear half-space constraints in LLM hidden space but requires per-category tuning of the constraint count K. We show th
arXiv:2607.19449v1 Announce Type: cross Abstract: Evaluation frameworks for tool-augmented LLM agents focus overwhelmingly on capability metrics or explicit tool crashes, leaving silent infrastructure
arXiv:2607.19490v1 Announce Type: cross Abstract: Peer-to-peer distributed inference executes a Large Language Model (LLM) on pooled consumer hardware by spreading its layers across many nodes. Every
arXiv:2604.03447v2 Announce Type: replace-cross Abstract: LLM-based software engineering assistants often reason over multiple artifacts, including code, documentation, signatures, and tests, even whe
arXiv:2607.19356v1 Announce Type: new Abstract: Tool-using LLM agents increasingly execute high-impact actions, making runtime safety monitoring essential. We present NEXUS (Neural EXecution Utility a
arXiv:2607.19824v1 Announce Type: new Abstract: LLM preference alignment aims to optimize models toward human preferences across diverse user instructions. Reinforcement learning has become a major po
arXiv:2607.19967v1 Announce Type: cross Abstract: Shippers are beginning to delegate carrier selection to large language model (LLM) agents. We ask what such delegation does to a freight matching mark
“Geologists communicated in English; and they could name things in a manner that sent shivers through the bones' Not a whiff of LLM feel, despite the em-dashes & period-separated lists. We focus too m
arXiv:2607.13347v1 Announce Type: cross Abstract: LLM-as-a-judge is widely used to provide feedback and selection signals in closedloop regeneration, but this use remains insufficiently validated. We
arXiv:2607.13511v1 Announce Type: cross Abstract: We introduce ExTernD (Expanded-rank Ternary Decomposition), a post-training factorization of each LLM weight matrix A in R^{m imes n} into A approx B
arXiv:2603.13952v3 Announce Type: replace-cross Abstract: In existing Audio-Visual Speech Enhancement (AVSE) methods, objectives such as Scale-Invariant Signal-to-Noise Ratio (SI-SNR) and Mean Squared
arXiv:2607.13918v1 Announce Type: cross Abstract: Serial verification gates are a core reliability primitive in LLM harnesses: a candidate answer is returned only if k verifier calls all accept it. Un
arXiv:2607.13618v1 Announce Type: new Abstract: LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the fin
arXiv:2607.13048v1 Announce Type: cross Abstract: Streaming inference pipelines increasingly pair lightweight fast models with Large Language Models (LLMs) that provide rich semantic understanding at
arXiv:2602.11619v2 Announce Type: replace Abstract: Running the same LLM agent on identical inputs yields 2.3-4.2 distinct action sequences per 10 runs; this behavioral variance constitutes a training
arXiv:2607.12468v1 Announce Type: cross Abstract: We describe our submission to Task 1 of the 2nd MLCSLM Challenge: a cascaded diarization-then-recognition system that combines DiariZen-Large-s80 (Wav
arXiv:2509.24372v3 Announce Type: replace-cross Abstract: Fine-tuning large language models (LLMs) for downstream tasks is an essential stage of modern AI deployment. Reinforcement learning (RL) has e
arXiv:2604.10925v2 Announce Type: cross Abstract: Natural language remains the predominant way people interact with large language models (LLMs). However, users often struggle to precisely express and
arXiv:2607.12621v1 Announce Type: new Abstract: Recent work has shown that 'Vision-Free'' approaches (representing images as text) can be effective for standard image retrieval tasks. However, it rema
Grok is the Flow LLM. Grok 4.5’s biggest advantage is its speed. It’s smart enough to be comparable to the other models on most things. But that speed allows you to make little tweaks to your system s
AI productivity is best measured by connecting AI usage to validated downstream outcomes. Tokens, prompts, and generated lines show activity, but they do not prove value. A better measurement model tr
🆕 In Code They Act, In Proof We Trust — Erik Meijer last year, @solomonstre defined agents as 'an LLM that's wrecking its environment in a loop', and @simonw coined the Lethal Trifecta for agents, tha
Hardmaru posted a tweet on July 13, 2026 stating that while language models and coding agents are valuable, “there is more to life, and more to AI, than just LLM agents.” The tweet received 61.1 k vie
1/ I watched LangChain’s webinar on LLM wikis and agent memory and wrote up a working paper from my own small agent-fleet experiments: Soft-Cache: A Human-Supervisable Coherence Protocol for Persisten
arXiv:2505.23842v5 Announce Type: replace Abstract: Large Language Models (LLMs) increasingly power search engines and AI assistants that retrieve and summarize content from many sources. By serving a
arXiv:2607.07321v1 Announce Type: new Abstract: Tool utilization enables Large Language Model (LLM) agents to interact with the real world and resolve complex tasks. However, existing agent frameworks
arXiv:2607.07494v1 Announce Type: cross Abstract: Gradient communication is a primary scaling bottleneck in large language model (LLM) pretraining. Communicating gradients in low-precision formats, su
arXiv:2607.07405v1 Announce Type: new Abstract: Tool-using LLM agents can violate the very policies they are deployed to enforce while appearing to complete the task successfully. In policy-permissive
Thank you @hwchase17 @BraceSproul @devstein64 @jeffreyhuber for the LLM Wikis session today, I absolutely loved it! Really grateful you keep these sessions open and honest about what actually works in
arXiv:2510.17476v2 Announce Type: replace Abstract: Equitable access to reliable health information is vital when integrating AI into healthcare. Yet, information quality varies across languages, rais
arXiv:2607.05708v1 Announce Type: new Abstract: Recent LLM-based agent systems continuously accumulate context across multi-turn interactions, tool invocations, and cross-session workflows. Replaying
arXiv:2607.05694v1 Announce Type: cross Abstract: Logit-based watermarking is a widely used mechanism for identifying LLM generated content, yet its effectiveness is governed by a fundamental trade-of
arXiv:2510.19771v4 Announce Type: replace Abstract: LLM-based agents are increasingly moving towards proactivity: rather than awaiting instruction, they exercise agency to anticipate user needs and so
arXiv:2607.06503v1 Announce Type: new Abstract: Large language model (LLM) agents solving multi-step tasks frequently commit to trajectories that are doomed to fail, yet continue to consume substantia
arXiv:2607.05970v1 Announce Type: cross Abstract: Dataset search depends heavily on metadata, making LLM-generated metadata a consequential form of synthetic content in retrieval systems. We study six
arXiv:2607.06519v1 Announce Type: new Abstract: Long-context LLM inference is increasingly limited by the memory and bandwidth cost of KV caches, yet aggressive compression can remove the layer-specif
arXiv:2607.06080v1 Announce Type: cross Abstract: Putnam's Social Capital Theory is a foundational framework for collective action and community prosperity. However, traditional empirical methods face
Highly-recommended read. Aligns with what I see in my own harness: > Pi harness got the same success rate as harnesses from the LLM vendors with Opus and GPT, but at 2x less cost > GLM 5.2 was a major
arXiv:2603.20182v4 Announce Type: replace Abstract: Although robot-to-robot (R2R) communication improves indoor scene understanding beyond what a single robot can achieve, R2R alone cannot overcome pa
arXiv:2607.06223v1 Announce Type: new Abstract: Reinforcement learning has become a promising paradigm for improving large language model (LLM) agents on long-horizon search tasks, where the agent mus
arXiv:2607.06157v1 Announce Type: cross Abstract: Deliberation plays a crucial role in collaboration; when humans work together, they naturally engage in communication to align information and reach a
arXiv:2507.22758v2 Announce Type: replace Abstract: Recent advancements in financial problem-solving have leveraged LLMs and agent-based systems, with a primary focus on trading and financial modeling
arXiv:2607.05904v1 Announce Type: new Abstract: Training a language model against its own reference-free judgments (the premise of self-rewarding, self-play, and LLM-as-a-judge pipelines) assumes a mo
arXiv:2607.05545v1 Announce Type: cross Abstract: LLM conformity is often used to describe cases where a model changes a correct answer toward a peer or group response. We show that most of this appar
arXiv:2607.05876v1 Announce Type: cross Abstract: LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the re
We're releasing ZML/LLMD, our homegrown LLM server built on top of our homegrown high performance heterogeneous inference stack. It ships with 5 architectures out of the box: NVIDIA, AMD, Metal, Intel
arXiv:2607.04572v1 Announce Type: new Abstract: Large language model (LLM) tutors often produce fluent step-by-step explanations, but a correct and pedagogically formatted response does not guarantee
arXiv:2607.04668v1 Announce Type: cross Abstract: On-device LLM decoding is a hard-barriered CPU-SIMD computation that wants every core for milliseconds per token, while the rest of the OS wants those
arXiv:2607.02574v1 Announce Type: cross Abstract: The key-value (KV) cache has become a first-order memory object in LLM serving rather than a temporary per-request tensor. This survey classifies more
arXiv:2607.04302v1 Announce Type: cross Abstract: We present HiFA4, a post-training operator-level design that executes both QK^T and PV in FlashAttention as 4-bit HIF4 Cube GEMMs for LLM inference on
arXiv:2607.05031v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used to produce test oracles, the part of a test that decides whether observed behavior is correct. Yet
llm wikis are a glimpse of the future of what agent memory looks like this blog i wrote resonated with a lot of folks will be discussing this (as well as some updates ive made to my beliefs since this
arXiv:2607.05297v1 Announce Type: new Abstract: Recent LLM agents tackle increasingly long-horizon, open-ended tasks, and external skills, reusable procedural knowledge supplied to the agent, further
arXiv:2607.02525v1 Announce Type: cross Abstract: We present PEEK, a lightweight scheduling and eviction framework for both online (streaming) and offline (batch) LLM serving; this paper focuses on th
arXiv:2601.11960v3 Announce Type: replace-cross Abstract: Existing reinforcement learning methods for LLM reasoning implicitly assume that the policy generating training trajectories should coincide w
arXiv:2512.15765v3 Announce Type: replace Abstract: Data valuation is a natural framework for understanding which preference datasets matter most when aligning a Large Language Model (LLM) using multi
arXiv:2607.03333v1 Announce Type: cross Abstract: LLM agents are becoming a common interface for research, coding, and question answering, yet their Thought-Action-Observation loop is often serial: th
arXiv:2607.04963v1 Announce Type: new Abstract: Reinforcement Learning (RL) is the dominant paradigm for training Large Language Model (LLM) agents on long-horizon tasks. However, sparse and delayed r