Knowledge catalogue
arXiv:2605.06355v2 Announce Type: replace Abstract: Order-Agnostic autoregressive models have demonstrated strong performance in deep generative modeling, yet their use in settings with incomplete dat
arXiv:2605.30148v1 Announce Type: cross Abstract: Evolution Strategies (ES) has recently emerged as a competitive alternative to reinforcement learning (RL) for large language model (LLM) fine-tuning,
arXiv:2605.29407v1 Announce Type: new Abstract: This paper presents a phase-conditioned, force-aware framework for robust deformable object manipulation. Standard imitation learning policies such as A
arXiv:2605.29299v1 Announce Type: cross Abstract: Evaluations of dental vision-language models remain fragmented across datasets, task definitions and metrics, and often ignore their computational cos
arXiv:2602.10520v3 Announce Type: replace Abstract: Looped Language Models (LoopLMs) perform multi-step latent reasoning prior to token generation and outperform conventional LLMs on reasoning benchma
arXiv:2601.22347v2 Announce Type: replace-cross Abstract: Recent post-training quantization (PTQ) methods have adopted block rotations to diffuse outliers prior to rounding. While this reduces the ove
arXiv:2605.29538v1 Announce Type: new Abstract: With the emergence of wireless applications in three-dimensional environments, such as the low-altitude airspace and 3D heterogeneous networks, radio ma
Ray Data LLM, developed by Anyscale, achieves 2x higher throughput compared to vLLM's synchronous LLM engine when operating at production scale. This performance improvement likely stems from Ray Data
arXiv:2605.29192v1 Announce Type: new Abstract: Chain-of-thought traces from large reasoning models can span tens of thousands of tokens, yet we lack a vocabulary for describing their internal structu
arXiv:2605.26108v2 Announce Type: replace Abstract: Recent advances in few-step diffusion distillation have enabled efficient image generation, yet aligning these models with human preferences remains
arXiv:2605.29397v1 Announce Type: new Abstract: HTML observations in LLM-based web agents are extremely long, and while many reduction methods have been proposed, it remains unclear which methods redu
arXiv:2601.18728v2 Announce Type: replace Abstract: Modern generative modeling methods have demonstrated strong performance in learning complex data distributions from clean samples. In many scientifi
arXiv:2605.30049v1 Announce Type: new Abstract: Diffusion Transformers have become a powerful backbone for text-to-image generation, but their layered and cross-modal generation process makes safety c
arXiv:2602.21565v2 Announce Type: replace Abstract: Generative Flow Networks (GFlowNets) learn to sample diverse candidates in proportion to a reward function, making them well-suited for scientific d
Step 3.7 Flash is a 198B-parameter Mixture-of-Experts vision-language model designed for enterprise-scale production workloads, featuring native image and video input, a 256k context window, and confi
arXiv:2605.28831v1 Announce Type: cross Abstract: Long-horizon interactive agents often accumulate large trajectory histories yet still fail to answer questions about earlier events reliably. We argue
arXiv:2605.29796v1 Announce Type: new Abstract: Agentic search enables LLMs to solve complex multi-hop questions through iterative reasoning and external search. Despite the effectiveness, these syste
arXiv:2509.21707v3 Announce Type: replace-cross Abstract: Semi-supervised learning (SSL) arises in practice when labeled data are scarce or expensive to obtain, while large quantities of unlabeled dat
arXiv:2605.30239v1 Announce Type: new Abstract: This work addresses the problem of recovering complete, simulatable object geometry from reconstructed real-world scenes, enabling physics-based interac
arXiv:2603.23853v3 Announce Type: replace Abstract: Combining multiple Vision-Language Models (VLMs) can enhance multimodal reasoning and robustness, but aggregating heterogeneous models' outputs ampl
arXiv:2605.29402v1 Announce Type: cross Abstract: Understanding long-form egocentric videos remains challenging for multimodal large language models (MLLMs) due to limited context length and insuffici
arXiv:2508.02537v3 Announce Type: replace Abstract: Physics-Informed Neural Networks (PINNs) offer a powerful framework for solving PDEs by embedding physical laws into the learning process. However,
arXiv:2605.29863v1 Announce Type: new Abstract: Predicting the next mobile application a user will launch is essential for intelligent device resource management and proactive assistance. Existing mod
arXiv:2601.10960v2 Announce Type: replace-cross Abstract: Controllable generation requires language models to realize output characteristics such as reading level, politeness, and toxicity. Existing s
arXiv:2605.29194v1 Announce Type: cross Abstract: Many stochastic physical systems evolve smoothly over time in the sense that the distribution of states changes regularly across time steps. The trans
arXiv:2512.03010v2 Announce Type: replace Abstract: LiDAR-captured point clouds are often considered the gold standard in active 3D reconstruction. While their accuracy is exceptional in flat regions,
TERAFAB IS GOING TO BE INSANE. We’re targeting 100–200 billion custom AI + memory chips per year at full ramp that’s 1 terawatt (1,000 GW) of annual AI compute capacity. Roughly 50x current global AI
Thanks for the call out and glad to see that our blog inspired the broader community! We at @FireworksAI_HQ have been using delta updates to scale RL globally for about a year. Including Composer 2 &
arXiv:2605.29123v1 Announce Type: new Abstract: Masked diffusion language models (MDMs) uniquely support any-order generation, with confidence-based decoding currently serving as the de facto standard
arXiv:2605.28983v1 Announce Type: cross Abstract: In this paper, training a neural network is identified, exactly, as a search through Hamilton--Jacobi initial-value problems: each gradient step selec
arXiv:2605.29493v1 Announce Type: cross Abstract: Since public access to generative AI tools became widespread, federal civil litigation has seen a marked increase in pro se (self-represented) plainti
arXiv:2603.23971v2 Announce Type: replace-cross Abstract: Developers and consumers increasingly choose reasoning models (RMs) based on their listed API prices. However, how accurately do these prices
arXiv:2605.30040v1 Announce Type: cross Abstract: Per-token billing is now the standard pricing model for commercial large language models (LLMs), so the honesty of reported token counts directly affe
arXiv:2605.30060v1 Announce Type: new Abstract: This work presents ViGeo, a feed-forward foundation model for recovering spatially dense and temporally consistent geometry from video sequences. Built
arXiv:2605.29292v1 Announce Type: new Abstract: This technical report presents our solution for the CVPR 2026 UG2+ Challenge Track 3: Dynamic Object Segmentation in Turbulence (DOST). We design a trai
arXiv:2605.29534v1 Announce Type: new Abstract: Recent advances in mobile GUI agents have shown strong potential for automating mobile tasks, but most effective systems still depend on large vision-la
arXiv:2605.05964v2 Announce Type: replace Abstract: Quantifying uncertainty in neural network predictions is essential for high-stakes domains such as autonomous driving, healthcare, and manufacturing
arXiv:2605.30011v1 Announce Type: cross Abstract: Recent work has begun to equip vision-language-action (VLA) policies with explicit intermediate reasoning. In embodied control, however, textual chain
arXiv:2605.30330v1 Announce Type: new Abstract: Diffusion models have excellent capacity to model complex distributions of natural data, which has made them a popular and effective choice for posterio
arXiv:2602.16610v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are increasingly applied as automatic evaluators for natural language generation assessment often using pairwise
arXiv:2604.04295v3 Announce Type: replace Abstract: Automated patent claim validation demands low error tolerance. However, existing approaches face a rigidity-resource dilemma: lightweight encoders c
arXiv:2605.28607v1 Announce Type: new Abstract: Modern information systems require autonomous agents capable of navigating complex workflows, yet current methodologies often struggle with the transiti
arXiv:2605.28145v1 Announce Type: new Abstract: We present an adaptive reservoir computing framework for the CTF-4-Science Lorenz benchmark, which evaluates machine learning models across twelve disti
arXiv:2601.10714v2 Announce Type: replace Abstract: We introduce Alterbute, a diffusion-based method for editing an object's intrinsic attributes in an image. We allow changing color, texture, materia
arXiv:2407.21075v2 Announce Type: replace Abstract: We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficie
arXiv:2605.28659v1 Announce Type: new Abstract: Biological foundation models have shown strong performance in single-cell representation learning by applying transformer architectures directly to gene
arXiv:2605.28809v1 Announce Type: new Abstract: Class-Incremental Learning (CIL) is important in building real-world learning systems. In CLIP-based CIL, the model performs classification by comparing
arXiv:2605.28313v1 Announce Type: new Abstract: Large Language Models (LLMs) have demonstrated remarkable capabilities in tasks related to reasoning and judgment. However, assessing the quality of arg
arXiv:2605.28642v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have demonstrated significant potential for speech-to-text translation (S2TT). However, existing deployment par
arXiv:2605.28441v1 Announce Type: cross Abstract: While Contrastive Learning (CL) has revolutionized self-supervised representation learning, its latent representations remain highly entangled and opa
arXiv:2601.18116v2 Announce Type: replace Abstract: We argue that multi-document reasoning is constrained not only by how much text a model can read, but also by how limited query-time evidence budget
arXiv:2605.28780v1 Announce Type: new Abstract: Vision classifiers can exploit spurious correlations, achieving high in-distribution accuracy yet failing under distribution shift. Existing approaches
arXiv:2605.28070v1 Announce Type: new Abstract: We highlight a failure mode of large reasoning models on questions with insufficient information: models may recognize that a problem is under-specified
arXiv:2605.28089v1 Announce Type: new Abstract: BuddyBench introduces a privacy-constrained multi-task benchmark for pediatric social-communication personalization. Unlike existing neurodevelopmental
arXiv:2605.27725v1 Announce Type: cross Abstract: High-fidelity computational fluid dynamics (CFD) is widely used for thermal-fluid design, but repeated CFD solves remain expensive for design optimiza
arXiv:2605.27805v1 Announce Type: cross Abstract: While LLMs enable personalized chatbots, their effectiveness in child-centered personalization remains unclear, as systematic evaluation of child-spec
arXiv:2605.28111v1 Announce Type: new Abstract: Predicting how a cell will change its transcriptional state under a developmental signal or a genetic perturbation is the computational core of in-silic
arXiv:2508.21046v3 Announce Type: replace Abstract: Recent Vision-Language-Action (VLA) models built on pre-trained Vision-Language Models (VLMs) require extensive post-training, resulting in high com
arXiv:2602.18982v4 Announce Type: replace Abstract: Common deep learning approaches for antibody engineering focus on modeling the marginal distribution of sequences. By treating sequences as independ
arXiv:2605.28763v1 Announce Type: new Abstract: Interactive 3D assets used in games and simulation are typically decomposed into specific semantic parts to support animation, physics, and scripted beh