All entries

arXiv:2605.04523v1 Announce Type: new Abstract: We present our winning system for Task~B (generation with reference passages) in SemEval-2026 Task~8: MTRAGEval. Our method is a heterogeneous ensemble

arXiv:2603.06542v2 Announce Type: replace-cross Abstract: Conversational generative AI is increasingly explored in healthcare, where models must integrate heterogeneous patient signals and support div

Ramp Labs is doing a ton of cool work all the way from launching products to hardcore research on LLM internals, this is a fun one 🔥 Talked to @ramplabs Head of Applied Research Alex Shevchenko on the

Replit announced a security center feature designed to help users monitor and manage security aspects of their projects and accounts. The security center likely provides tools for vulnerability manage

arXiv:2605.03491v1 Announce Type: new Abstract: Most evaluations of autonomous driving policies under adversarial conditions are conducted in simulation, due to cost efficiency and the absence of phys

NCCL Inspector is a profiling plugin that provides detailed, per-communicator, per-collective performance and metadata logging, designed to help users analyze and debug NCCL collective operations by g

arXiv:2307.03017v5 Announce Type: replace Abstract: With the rise of Extended Reality (XR) technology, there is a growing need for real-time light field reconstruction from sparse view inputs. Existin

arXiv:2605.03361v2 Announce Type: new Abstract: As multimodal content continues to expand at a rapid pace, audio retrieval has emerged as a key enabling technology for media search, content organizati

arXiv:2508.04204v2 Announce Type: replace Abstract: Large Reasoning Models (LRMs) have demonstrated impressive performance in reasoning-intensive tasks, but they remain vulnerable to harmful content g

arXiv:2605.04019v1 Announce Type: new Abstract: AI systems are entering critical domains like healthcare, finance, and defense, yet remain vulnerable to adversarial attacks. While AI red teaming is a

arXiv:2605.05053v1 Announce Type: cross Abstract: Tactile perception is key to dexterous manipulation, yet simulating high-resolution elastomer deformation remains computationally prohibitive. Finite

arXiv:2605.04606v1 Announce Type: new Abstract: Traditional one-shot detection methods have addressed the closed-set problem in object detection, but the high cost of data annotation remains a critica

arXiv:2605.04647v1 Announce Type: new Abstract: We introduce ReflectDrive-2, a masked discrete diffusion planner with separate action expert for autonomous driving that represents plans as discrete tr

arXiv:2605.04895v1 Announce Type: new Abstract: Published transfer-BO comparisons often estimate an average treatment effect of acquisition choice over hidden regime variables, while practitioners nee

arXiv:2605.04920v1 Announce Type: cross Abstract: Compositional generalization refers to correctly interpret novel combinations of known primitives, which remains a major challenge. Existing approache

arXiv:2605.04965v1 Announce Type: new Abstract: Optimal transport (OT) is a central framework for modeling distribution shifts. Because OT compares distributions directly in input space, a well-design

arXiv:2605.04451v1 Announce Type: new Abstract: Geospatial reasoning requires models to resolve complex spatial semantics and user intent into precise target locations for Earth observation. Recent pr

arXiv:2605.03426v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have broad potential in privacy-sensitive domains such as healthcare and finance, yet strict data-sharing constraints rend

arXiv:2605.04832v1 Announce Type: new Abstract: Neural operators generally demonstrate strong predictive performance on in-distribution (ID) problems. However, a critical limitation of existing method

Research from Prof Julian Togelius found that despite AI's well-documented victories in chess, Go, and Atari games, humans still learn unfamiliar video games far faster than any AI model. #NYUTandonMa

Andy Greenberg / Wired: Researchers: 5,000+ web apps built using AI coding tools like Lovable, Base44, and Replit have little to no authentication, and ~40% exposed sensitive data — Companies like Lov

arXiv:2605.04075v1 Announce Type: cross Abstract: Multimodal Large Language Models face severe challenges in computational efficiency and memory consumption due to the substantial expansion of the vis

arXiv:2605.04723v1 Announce Type: cross Abstract: Attribute-aware sequential recommendation entails predicting the next item a user will interact with based on a chronologically ordered history of pas

arXiv:2605.04913v1 Announce Type: new Abstract: LLM post-training typically propagates task gradients through the full depth of the model. Although this end-to-end structure is simple and general, it

arXiv:2603.13783v2 Announce Type: replace Abstract: Temporal retiming, the ability to reconstruct and render dynamic scenes at arbitrary timestamps, is crucial for applications such as slow-motion pla

arXiv:2605.03308v1 Announce Type: new Abstract: Travel planning serves as a critical task for long-horizon reasoning, exposing significant deficits in LLMs. However, existing benchmarks and evaluation

arXiv:2605.04531v1 Announce Type: new Abstract: Open-vocabulary object detection with vision-language models (VLMs) such as Grounding DINO suffers from performance degradation under test-time distribu

arXiv:2605.04607v1 Announce Type: new Abstract: This paper presents a multi-phase whole-body model predictive control approach for bipedal walking, combining a detailed whole-body model in the near ho

arXiv:2605.04539v1 Announce Type: new Abstract: Direct Preference Optimization (DPO), the efficient alternative to PPO-based RLHF, falls short on knowledge-intensive generation: standard preference si

arXiv:2605.04242v1 Announce Type: new Abstract: Nationwide road-incident forecasting is a systems problem before it is a modeling problem. A usable service must connect historical incident archives, h

Researchers developed a control method for Reachy Mini robots that uses a real human face as an input controller rather than building a humanoid facial interface. This approach leverages existing robo

arXiv:2605.03409v1 Announce Type: new Abstract: We present Robust Agent Compensation (RAC), a log-based recovery paradigm (providing a safety net) implemented through an architectural extension that c

ROCm has reached production-ready status for PyTorch and vLLM workloads in 2026. PyTorch runs well and MI300X benchmarks are competitive, though the ecosystem gap with CUDA remains real. AMD's MI355X

arXiv:2412.03077v2 Announce Type: replace Abstract: 4D reconstruction from casually captured monocular videos is challenging due to inherent ambiguity in reconstructing dynamic 3D geometry. To address

arXiv:2605.05112v1 Announce Type: new Abstract: SWE-bench-style agentic reinforcement learning relies on expensive stateful trajectories, yet substantial compute is wasted on sampled rollout groups wi

arXiv:2504.05407v2 Announce Type: replace-cross Abstract: Autonomous surveillance missions in Internet of Things (IoT) networks often involve solving NP-hard combinatorial optimization problems to ens

arXiv:2511.19513v2 Announce Type: replace Abstract: Decentralized learning often involves a weighted global loss with heterogeneous node weights lambda. We revisit two natural strategies for incorpora

arXiv:2601.00264v2 Announce Type: replace Abstract: Multimodal learning has revolutionized general domain tasks, yet its application in scientific discovery is hindered by the profound semantic gap be

arXiv:2601.08623v2 Announce Type: replace Abstract: Image generation models (IGMs), while capable of producing impressive and creative content, often memorize a wide range of undesirable concepts from

arXiv:2605.04222v1 Announce Type: cross Abstract: Real-world control systems must achieve long-horizon objectives (liveness) while respecting continuous-time safety constraints, a combination that mot

arXiv:2502.04512v3 Announce Type: replace Abstract: AI advancements have been significantly driven by a combination of foundation models and curiosity-driven learning aimed at increasing capability an

日経デジタルガバナンスにて、Sakana AIの「事後学習」技術と Sakana Chat / Namazuモデルに関する記事が掲載されました。 Namazuモデル開発をリードしたリサーチサイエンティストの菅沼雅徳とチーフサイエンティストの秋葉拓哉に取材いただきました。 記事: https://www.nikkei.com/prime/digital-governance/article/DGXZ

arXiv:2605.02897v1 Announce Type: cross Abstract: LLM assistant personalities play a critical role in user experience and perceived response quality. We present a large-scale experiment of frontier LL

arXiv:2605.04560v1 Announce Type: new Abstract: Perceptual image compression focuses on preserving high visual quality under low-bitrate constraints. Most existing approaches to perceptual compression

Saw this and thought 'yes! ChatGPT voice mode is going to stop acting like a two-year-model' but that upgrade hasn't shipped just yet Introducing GPT-Realtime-2 in the API: our most intelligent voice

arXiv:2605.05008v1 Announce Type: cross Abstract: Regionalization aims to partition a spatial domain into contiguous regions that share similar characteristics, enabling more effective spatial analysi