Knowledge catalogue
Langsmith Engine appears to be a seriously high value release from the @LangChain team A deep agent that’s monitoring Langsmith agent traces and not only surfaces issues but: - proposes a PR to resolv
arXiv:2605.11348v1 Announce Type: new Abstract: During disasters, extracting causal relations from social media can strengthen situational awareness by identifying factors linked to casualties, physic
arXiv:2502.12276v2 Announce Type: replace Abstract: Computational methods for analyzing prose and poetry utilize word embeddings and other abstract representations that sometimes obscure context-rich
arXiv:2501.02955v2 Announce Type: replace Abstract: In recent years, vision language models (VLMs) have made significant advancements in video understanding. However, a crucial capability - fine-grain
arXiv:2411.16769v3 Announce Type: replace-cross Abstract: Understanding the capabilities of text-to-image (T2I) models in harmful content generation is essential to safety and compliance. However, hum
arXiv:2508.10036v2 Announce Type: replace Abstract: Large Language Models (LLMs) show remarkable potential for few-shot information extraction (IE), yet their performance is highly sensitive to the ch
arXiv:2605.11242v1 Announce Type: new Abstract: In this paper, we present the RETUYT-INCO participation at the BEA 2026 shared task 'Rubric-based Short Answer Scoring for German'. Our team participate
arXiv:2605.10973v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) improves in-domain performance but can degrade out-of-domain (OOD) generalization. Prior work suggests that this degradatio
arXiv:2605.12163v1 Announce Type: new Abstract: In language reasoning, longer chains of thought consistently yield better performance, which naturally suggests that visual latent reasoning may likewis
arXiv:2605.11246v1 Announce Type: new Abstract: Offline black-box optimization aims to discover novel designs with high property scores using only a static dataset, a task fundamentally challenged by
arXiv:2605.11374v1 Announce Type: cross Abstract: Test-time compute is widely believed to benefit only large reasoning models. We show it also helps small embedding models. Most modern embedding check
arXiv:2507.13841v2 Announce Type: replace Abstract: Good storytelling involves surprise -- unpredictability in how the story unfolds -- and sense-making, the requirement that the story forms a coheren
arXiv:2605.11864v1 Announce Type: cross Abstract: Listwise reranking is a key yet computationally expensive component in vision-centric retrieval and multimodal retrieval-augmented generation (M-RAG)
We built SmithDB: the database purpose built for agent observability workloads that now powers many parts of LangSmith. Agent observability presents a challenging data problem. Agent traces can contai
arXiv:2605.11612v1 Announce Type: new Abstract: Backdoor vulnerabilities widely exist in the fine-tuning of large language models(LLMs). Most backdoor poisoning methods operate mainly at the token lev
arXiv:2605.11240v1 Announce Type: cross Abstract: Generative AI models differ from traditional machine learning tools in that they allow users to provide as much or as little information as they choos
arXiv:2605.10204v1 Announce Type: new Abstract: Accurate 3D reconstruction of objects with reflective, transparent, or low-texture surfaces still remains notoriously challenging. Such materials often
arXiv:2605.08432v1 Announce Type: cross Abstract: Calibration measures whether a model's predicted confidence aligns with its empirical accuracy, and is central to the reliable deployment of large lan
arXiv:2605.08504v1 Announce Type: new Abstract: We investigate the origins of massive activations in large language models (LLMs) and identify a specific layer named the extbf{Massive Emergence Layer
arXiv:2605.09034v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization has become increasingly popular and important in fine-tuning large language models (LLMs), especially on edge devices due
arXiv:2605.10365v1 Announce Type: new Abstract: Autonomous agents have rapidly matured as task executors and seen widespread deployment via harnesses such as OpenClaw. Safety concerns have rightly dra
arXiv:2605.08401v1 Announce Type: cross Abstract: Recent advances in large language models (LLMs) have demonstrated remarkable reasoning capabilities, largely stimulated by Reinforcement Learning with
arXiv:2605.09698v1 Announce Type: new Abstract: As data-science agents shift from co-pilots to auto-pilots, silent misframing becomes a critical failure mode. Agents quietly commit to plausible but un
arXiv:2605.10339v1 Announce Type: new Abstract: The advancement of Large Language Models (LLMs) has enabled their application in personalized dialogue systems. We present an extended annotation scheme
arXiv:2605.10107v1 Announce Type: new Abstract: Assertion-based Verification (ABV) is essential for ensuring that hardware designs conform to their intended specifications. However, existing automated
arXiv:2502.09891v4 Announce Type: replace-cross Abstract: Retrieval-Augmented Generation (RAG) has proven effective in integrating external knowledge into large language models (LLMs) for solving ques
arXiv:2512.15977v3 Announce Type: replace Abstract: Vision-language models (VLMs) are increasingly proposed as general-purpose solutions for visual recognition tasks, yet their reliability for agricul
B9124 is a build release from the llama.cpp project, created on May 12, 2026 . Llama.cpp is an open source software library that performs inference on various large language models such as Llama, deve
arXiv:2605.10845v1 Announce Type: cross Abstract: As global cross-lingual communication intensifies, language barriers in visually rich documents such as PDFs remain a practical bottleneck. Existing d
arXiv:2605.08463v1 Announce Type: new Abstract: Autonomous AI agents are increasingly deployed in open social environments, yet the relationship between their configuration specifications and their em
As AI coding agents become deeply embedded in developer workflows, defenders must evolve their definition of malicious files and rethink how to protect against them. Autonomous AI agents operate acros
arXiv:2605.08558v1 Announce Type: new Abstract: As an extension of the classical multi-armed bandit problem, multi-fidelity multi-armed bandits (MF-MAB) enable individual arms to be evaluated using di
arXiv:2502.08943v4 Announce Type: replace-cross Abstract: Large language models (LLMs) have demonstrated significant utility in real-world applications, exhibiting impressive capabilities in natural l
arXiv:2512.24863v2 Announce Type: replace-cross Abstract: The world is in the grip of ecological, meaning, and language crises that are converging into a metacrisis. Big AI is accelerating them all. L
arXiv:2601.12369v3 Announce Type: replace Abstract: Deep Research Agents increasingly automate survey generation, yet whether they match human experts at retrieving essential papers and organizing the
arXiv:2605.09223v1 Announce Type: new Abstract: Understanding long videos with multimodal large language models (MLLMs) requires selecting a small subset of informative frames under strict computation
arXiv:2603.02678v3 Announce Type: replace Abstract: This paper argues for recognizing an emerging paradigm of causal learning by wisdom of the crowd. Recent developments in government, industry, and r
arXiv:2605.09079v1 Announce Type: new Abstract: Despite surpassing human performance across mathematics, coding, and other knowledge-intensive tasks, large language models (LLMs) continue to struggle
arXiv:2605.10362v1 Announce Type: new Abstract: Training AI models for computational pathology currently requires access to expensive whole-slide-image datasets, GPU infrastructure, deep expertise in
arXiv:2603.28902v2 Announce Type: replace Abstract: Charts are central to analytical reasoning, yet existing benchmarks for chart understanding focus almost exclusively on single-chart interpretation
arXiv:2605.09168v1 Announce Type: new Abstract: A valid tool call is not necessarily a valid intervention. Tool-using language agents are guarded by schema validators, policy filters, provenance check
🤩🤯🤩 Claude Code (still not AGI but biggest advance since GPT-4) is the most neurosymbolic thing I have ever seen in my life. 53 symbolic tools, 500,000 lines of symbolic code, combined with a state-of
arXiv:2509.14234v3 Announce Type: replace Abstract: Where do learning signals come from when there is no ground truth in post-training? We show that inference compute itself can serve as supervision.
arXiv:2605.09045v1 Announce Type: new Abstract: Agentic frameworks are the software layer through which AI agents act in the world. Existing safety methods intervene on the model and therefore remain
Cool idea from Nous Research. What if you could speed up long-context pretraining with a subquadratic wrapper that you remove before deployment? That is the idea behind Lighthouse Attention. The metho
arXiv:2602.05902v2 Announce Type: replace-cross Abstract: Post-training quantization (PTQ) enables efficient deployment of large language models by mapping pretrained weights to low-bit formats withou
arXiv:2605.08389v1 Announce Type: cross Abstract: Zero-shot composed image retrieval (ZS-CIR) retrieves a target image from a reference image and a text modification without human-annotated CIR triple
Deep learning hit a wall. Neurosymbolic AI rescued it. 🤩🤯🤩 Claude Code (still not AGI but biggest advance since GPT-4) is the most neurosymbolic thing I have ever seen in my life. 53 symbolic tools, 5
arXiv:2512.06673v2 Announce Type: replace Abstract: Multimodal large language models (MLLMs) are rapidly expanding from general video understanding to finer-grained understanding such as spatio-tempor
arXiv:2605.10765v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, yet real-world deployment often requires continual cap
arXiv:2605.08695v1 Announce Type: new Abstract: Forensic analysis of AI-edited images requires more than binary real-versus-fake prediction: a useful system should localize the edit, identify its sema
arXiv:2605.09588v1 Announce Type: cross Abstract: Organizations increasingly deploy multiple AI systems across task domains, but selecting a small, high-performing ensemble can require costly model ca
arXiv:2601.18061v3 Announce Type: replace Abstract: Learning from human feedback~(LHF) assumes that expert judgments, appropriately aggregated, yield valid ground truth for training and evaluating AI
arXiv:2605.09853v1 Announce Type: new Abstract: Post-training techniques combined with inference-time scaling significantly enhance the reasoning and alignment capabilities of large language models (L
arXiv:2605.09438v1 Announce Type: new Abstract: Many features in pretrained Transformers span multiple layers: they emerge through stages of inference, persist in the residual stream, or are built joi
arXiv:2605.08140v1 Announce Type: cross Abstract: The Karlsruhe Tritium Neutrino Experiment (KATRIN) aims to measure the absolute neutrino mass with unprecedented sensitivity, requiring precise monito
arXiv:2605.08896v1 Announce Type: cross Abstract: Robust adaptation of LLMs and VLMs is often evaluated by average accuracy or average consistency under perturbations. However, these averages can hide
arXiv:2605.10834v1 Announce Type: new Abstract: AI pentesting agents are increasingly credible as offensive security systems, but current benchmarks still provide limited guidance on which will perfor
arXiv:2605.08838v1 Announce Type: cross Abstract: Retrieval-augmented generation (RAG) is widely used to augment large language models (LLMs) with external knowledge. However, many benchmark datasets,
arXiv:2604.07558v2 Announce Type: replace-cross Abstract: Digital mental health (DMH) tools have extensively explored personalization of interventions to users' needs and contexts. However, this perso