All entries

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

We just shipped tons of new products to accelerate the full agent development lifecycle: https://www.langchain.com/blog TLDR: ✅ LangSmith Engine ✅ SmithDB ✅ Sandboxes ✅ Managed Deep Agents ✅ LLM Gatew

🎉 We published a new AI safety study: shopping agents fall for whimsical attacks and lose money. A whimsical attack is an absurd scenario a human would never try on another human. In one run, GPT-5.1

arXiv:2605.11654v1 Announce Type: new Abstract: Cross-view geo-localization (CVGL), which matches an oblique drone view to a geo-referenced satellite tile, has emerged as a key alternative for autonom

arXiv:2502.11981v3 Announce Type: replace Abstract: Decades of research in machine learning have given us powerful tools for making accurate predictions. But when used in social settings and on human

We're building a database 🤠 We built SmithDB: the database purpose built for agent observability workloads that now powers many parts of LangSmith. Agent observability presents a challenging data prob

🔴 We're LIVE from the SaaStr floor, Day 2! 🎙️🚀 Joined by an amazing lineup today: • Lindsay Wise + Julia Holm (Green Security) • Kody Low (Replit Field Engineering) • Philipp Dietz (Replit Brand & Cre

We’re training models wrong and it’s due to chatGPT. Even the modern coding agents used daily still use message-based exchanges: They send messages to users, to themselves (CoT) and to tools, and rece

arXiv:2605.11963v1 Announce Type: new Abstract: This paper examines what it means for a medical AI system to be right by grounding the question in a specific clinical context: the automatic classifica

arXiv:2605.12281v1 Announce Type: new Abstract: What makes a word difficult to learn, and how does the difficulty depend on the learner's native language? We computationally model vocabulary difficult

What we find most useful about TST is the decoupling. The training-time efficiency is fully separated from the inference-time architecture, which makes TST a clean addition on top of other pretraining

arXiv:2605.12021v1 Announce Type: new Abstract: Many image understanding tasks involve identifying what is present and where it appears. However, tasks that address where, such as object discovery, de

arXiv:2605.11008v1 Announce Type: new Abstract: While invariant architectures are standard for processing symmetric data, there is growing interest in achieving invariance by applying group averaging

arXiv:2605.11867v1 Announce Type: new Abstract: Structural MRI-to-amyloid PET synthesis has been proposed as a non-invasive alternative for amyloid assessment in Alzheimer's disease (AD). However, rep

arXiv:2605.06314v2 Announce Type: replace Abstract: Benign overfitting is well-characterized in ell_2 geometries, but its behavior under the ell_1 implicit bias of greedy ensembles remains challenging

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:2605.11559v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have become a key interface for visual reasoning and grounded question answering, yet they remain vulnerable to

arXiv:2605.12112v1 Announce Type: new Abstract: RLHF is widely used to align flow-matching text-to-image models with human preferences, but often leads to severe diversity collapse after fine-tuning.

arXiv:2512.17738v2 Announce Type: replace Abstract: User-generated content (UGC) is characterised by frequent use of non-standard language, from spelling errors to expressive choices such as slang, ch

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

When you want to move from single agent SQLite on something like QMD, PostgreSQL is a great choice for multi agent and production quality, but not as snappy. So we made it much more snappy with BM25 &

arXiv:2605.12255v1 Announce Type: cross Abstract: When people share the same documents and observations yet reach different conclusions, the disagreement often shifts into a judgment that the other pa

💡 Why did @togethercompute choose NVIDIA Blackwell to serve DeepSeek-V4? Because NVIDIA Blackwell is built for the bottlenecks that matter most in long-context inference: → KV-cache pressure during de

arXiv:2605.11696v1 Announce Type: new Abstract: Recent single-image relighting methods, powered by advanced generative models, have achieved impressive photorealism on synthetic benchmarks. However, t

DeepSeek-v4 Flash through Ollama is currently available as a cloud model, where Ollama's CLI sends API calls to Ollama's hosted version rather than running locally . Local support for DeepSeek V4 Flas

arXiv:2605.12090v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have achieved strong semantic generalization for embodied policy learning, yet they learn reactive observation-to-

arXiv:2605.11743v1 Announce Type: new Abstract: Learning latent representations that capture both semantic and spatial information is central to efficient spatio-semantic reasoning. However, many exis

arXiv:2605.12162v1 Announce Type: new Abstract: Effectively handling the interplay between spatial perception and action generation remains a critical bottleneck in robotic manipulation. Existing meth

arXiv:2605.10981v1 Announce Type: new Abstract: Reference-free preference optimization has emerged as an efficient alternative to reinforcement learning from human feedback, with Simple Preference Opt

arXiv:2605.11521v1 Announce Type: new Abstract: Autonomous driving and intelligent transportation systems remain vulnerable under extreme weather. The U.S. Federal Highway Administration reports that

Yann LeCun says you cannot build a reliable agentic system without a world model LLMs don't have world models. They can't predict the consequences of their actions before taking them 'they just act, a

arXiv:2605.11906v1 Announce Type: new Abstract: Preference optimization has become an important post-training paradigm for improving the reasoning abilities of large language models. Existing methods

This article argues that infrastructure quality and architecture are more critical to AI/ML project success than the choice of underlying model. It likely explores how proper deployment, scaling, moni

arXiv:2510.02043v2 Announce Type: replace Abstract: Pose estimation refers to tracking a human's full body posture, including their head, torso, arms, and legs. The problem is challenging in practical

arXiv:2605.11435v1 Announce Type: new Abstract: In this paper, we propose a zero-reference diffusion-based framework, named ZeroIDIR, for illumination degradation image restoration, which decouples th

Katherine Bindley / Wall Street Journal: ZipRecruiter: IT and CS job postings are up 14.2% YoY in April; entry-level roles fell from 8.1% a year ago to 7.4%, while senior roles grew from 38.8% to 43.1

Nous Research has made Step 3.5 Flash, a faster AI model variant, available again on their portal at no cost for a 15-day period following popular demand after an initial 10-day availability window. T

arXiv:2605.08600v1 Announce Type: new Abstract: We present a new publicly available corpus of 100,502 movie reviews from Kazakhstan collected from kino.kz, spanning 2001-2025 and covering 4,943 unique

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:2511.00560v2 Announce Type: replace Abstract: Although 3D Gaussian Splatting (3D-GS) achieves efficient rendering for novel view synthesis, extending it to dynamic scenes still results in substa

The enterprise artificial intelligence stack is getting smarter — but is it getting the context it needs to make agentic AI deployment actually work? The defining problem of the agentic era might neve

A beginner-friendly guide to building a simple RAG (RETRIEVAL-AUGMENTED GENERATION) AI Agent with Pinecone + n8n 🧵 If you’re learning AI automation and you’re confused about Pinecone, knowledge bases,

arXiv:2605.08207v1 Announce Type: new Abstract: Pathology foundation models have shown strong retrospective performance, but whether such systems can support clinically relevant use remains unclear. T

arXiv:2605.08550v1 Announce Type: new Abstract: The population dynamics of molecules, cells, and organisms are governed by a number of unknown forces. In the last decade, population dynamics have pred

arXiv:2605.10086v1 Announce Type: new Abstract: This paper proposes a cell decomposition algorithm for binary occupancy grids that ensures mutual complete visibility from each cell to at least one adj

arXiv:2605.09483v1 Announce Type: cross Abstract: In this (work in progress) paper, we present Bounded Pragmatic Listener (or BPL), a cognitively grounded Bayesian framework for modelling susceptibili