Knowledge catalogue
This post appears to be a community prompt-sharing discussion where Swyx invites people to share their '/goal' prompts (likely system prompts or directives designed to optimize AI behavior toward spec
arXiv:2605.16311v1 Announce Type: new Abstract: Distributed training of large neural networks is bottlenecked by full-precision gradient communication and by coordinatewise optimizers that ignore the
arXiv:2512.04746v2 Announce Type: replace-cross Abstract: Extremely low-bit quantization is critical for efficiently deploying Large Language Models (LLMs), yet it often leads to severe performance de
arXiv:2605.17850v1 Announce Type: cross Abstract: iffusion-based generative models increasingly rely on inference-time guidance, adding a drift term or reweighting mixture of experts, to improve sampl
arXiv:2312.03798v2 Announce Type: replace Abstract: Single Image Reflection Removal (SIRR) in real-world images is a challenging task due to diverse image degradations occurring on the glass surface d
arXiv:2605.17341v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have achieved remarkable success, yet their reliance on massive datasets and unintended memorization of training data ra
arXiv:2505.21893v3 Announce Type: replace-cross Abstract: Preference learning has garnered extensive attention as an effective technique for aligning diffusion models with human preferences in visual
arXiv:2605.18221v1 Announce Type: cross Abstract: Real-time magnetic resonance imaging (rtMRI) of speech production enables non-invasive visualization of dynamic vocal-tract motion and is valuable for
arXiv:2605.16842v1 Announce Type: new Abstract: Diffusion Multi-Modal Large Language Models (dMLLMs) are powerful for image generation, but optimizing them through reinforcement learning (RL) remains
arXiv:2605.16650v1 Announce Type: cross Abstract: Evaluating multi-turn dialogue systems remains challenging because response quality depends not only on the current prompt, but also on previously est
arXiv:2605.18693v1 Announce Type: new Abstract: As LLM agents are increasingly built around reusable skills, a central challenge is no longer only whether agents can use provided skills, but whether t
arXiv:2602.14211v2 Announce Type: replace-cross Abstract: Agent skills are increasingly used to extend LLM agents with task-specific instructions, executable scripts, and auxiliary resources. While im
arXiv:2605.16986v1 Announce Type: cross Abstract: LLM agents benefit from reusable skills, yet test-time tasks often require guidance more specific than a static skill library can provide. We propose
arXiv:2605.18401v1 Announce Type: cross Abstract: Long-horizon LLM agents leave traces that could become reusable experience, but raw trajectories are noisy and hard to govern. We treat Agent Skills a
arXiv:2605.16565v1 Announce Type: new Abstract: Skim is a speculative execution framework for web agents that exploits the predictable structure of purpose-built websites. Today's web-agent expense is
arXiv:2605.17949v1 Announce Type: new Abstract: Remote sensing vision-language models commonly rely on pretrained visual encoders to convert images into semantic features before language-model reasoni
arXiv:2605.16626v1 Announce Type: cross Abstract: Since autonomous coding agents generate complex behaviors at high-volume, we may want to use other LLMs to monitor actions to reduce the risk from dan
arXiv:2604.08432v2 Announce Type: replace-cross Abstract: Photonic neural networks promise ultrafast inference, yet most architectures rely on linear optical meshes with electronic nonlinearities, rei
Sneaky interventions, interactive installations, tools built for other artists to use. @kcimc's practice lives where machine learning, computer vision, and social technology collide. See his new inter
arXiv:2605.17842v1 Announce Type: new Abstract: Autoregressive language models execute Transformer layers sequentially, creating a latency bottleneck that is not removed by conventional tensor or pipe
Google is transitioning from the Gemini CLI tool to a new CLI called AGY (Antigravity CLI), but this change has drawbacks: AGY is not open source and no longer supports ACP functionality, representing
arXiv:2605.17423v1 Announce Type: new Abstract: We study series-level cinematic remaking, a long-horizon video-to-video generation problem that localizes full episodes or films via stylization or acto
arXiv:2605.17789v1 Announce Type: cross Abstract: Memory systems for AI assistants were built for single-user dialogue and fail characteristically when applied to multi-party social group settings. Th
Cybersecurity startup Sola Security Ltd. today announced the launch of Lumina, an autonomous risk intelligence platform that applies contextual artificial intelligence across cloud, identity, software
arXiv:2605.18232v1 Announce Type: cross Abstract: Somali is a Cushitic language of the Horn of Africa with ~25 million speakers, yet no documented dedicated Somali pretraining corpus with a companion
This X thread from Google AI showcases practical and creative applications of Gemini Omni, Google's multimodal AI model, as demonstrated and shared by the user community. The thread appears to be a cu
arXiv:2605.16872v1 Announce Type: cross Abstract: AI agents increasingly act consequentially in the real world. This creates a problem we call consequence reception: harm occurs, the producing system
arXiv:2605.17710v1 Announce Type: new Abstract: Although modern multilingual Automatic Speech Recognition (ASR) systems support several Nigerian languages, their performance consistently lags behind h
arXiv:2511.00392v2 Announce Type: replace-cross Abstract: Accurate 3D reconstruction in visually-degraded underwater environments remains a formidable challenge. Single-modality approaches are insuffi
Bloomberg: Sources: Chinese AI startup Moonshot told investors it would revamp its corporate structure to pave the way for a Hong Kong IPO and comply with Beijing's rules — Moonshot AI has informed it
Financial Times: Sources: Demis Hassabis was an angel investor in Anthropic; PitchBook and Dealroom: former DeepMind researchers founded 12+ companies since 2021, raising $14B+ — Demis Hassabis, the f
Bloomberg: Sources detail growing concerns inside SoftBank over Masayoshi Son's 60B+ bet on OpenAI, which some fear concentrates too much capital into a single company — With more than 60 billion comm
Nikkei Asia: Sources: Intel asks its leading PC partners, including those in the US, China, and Taiwan, to use more 18A CPUs, citing better supply than chips on older nodes — TAIPEI — Intel is urging
Bloomberg: Sources: SpaceX expects to proceed with its acquisition of Cursor 30 days after its public trading debut, which is expected to occur on June 12 — SpaceX expects to proceed with its acquisit
Anna Tong / Forbes: Sources: Zyphra, which trains and runs inference for its open-weight models on AMD hardware, is raising a 500M Series B at a valuation of at least 5B — The Series B round, which ch
arXiv:2511.16309v2 Announce Type: replace Abstract: Sparse autoencoders (SAEs) are used to analyze embeddings, but their role and practical value are debated. We propose a new perspective on SAEs by d
arXiv:2509.23068v2 Announce Type: replace-cross Abstract: Recent advances in deep learning highlight the need for personalized models that can learn from small samples, handle high-dimensional feature
arXiv:2605.17156v1 Announce Type: cross Abstract: Quantum error correction (QEC) is essential for building fault-tolerant quantum computers, requiring decoders that are simultaneously accurate, fast,
arXiv:2505.19155v2 Announce Type: replace-cross Abstract: Due to the auto-regressive nature of current video large language models (Video-LLMs), the inference latency increases as the input sequence l
arXiv:2602.03535v2 Announce Type: replace Abstract: We introduce a dynamic sparse training algorithm based on linearized Bregman iterations / mirror descent that exploits the naturally incurred sparsi
arXiv:2605.17633v1 Announce Type: cross Abstract: The Segment Anything Model (SAM) achieves strong open-vocabulary segmentation, but its ViT-based image encoders dominate inference latency and memory.
arXiv:2605.17449v1 Announce Type: cross Abstract: Whole-slide MIL models are often called context-aware once graphs, Transform ers, or state-space modules are placed above patch embeddings. We show th
arXiv:2510.23641v2 Announce Type: replace-cross Abstract: Transformers are very effective in capturing both global and local correlations within high-energy particle collisions, but they present deplo
arXiv:2605.18209v1 Announce Type: cross Abstract: Spatial question answering over egocentric video is a challenging task that requires Vision-Language Models (VLMs) to reason about 3D object positions
arXiv:2603.29868v2 Announce Type: replace Abstract: The reliability of autonomous systems depends on their robustness, i.e., their ability to meet their objectives under uncertainty. In this paper, we
arXiv:2509.22061v2 Announce Type: replace-cross Abstract: Speech Continuation (SC) is the task of generating a coherent extension of a spoken prompt while preserving both semantic context and speaker
arXiv:2605.17311v1 Announce Type: new Abstract: The remarkable visual fidelity of recent commercial video generative models, such as Sora and Veo, renders robust AI-generated video detection increasin
arXiv:2605.18736v1 Announce Type: new Abstract: Diffusion models have been shown to implicitly generate visual content autoregressively in the frequency domain, where low-frequency components are gene
arXiv:2605.18466v1 Announce Type: new Abstract: Segmenting vocal tract articulators in real-time MRI (rtMRI) is a challenging dynamic image segmentation problem characterized by low contrast, rapid mo
arXiv:2601.09413v2 Announce Type: replace-cross Abstract: We introduce a voice-agentic framework that learns one critical omni-understanding skill: knowing when to trust itself versus when to consult
arXiv:2605.18389v1 Announce Type: new Abstract: Optimal transport provides a powerful framework for comparing measures while respecting the geometry of their support, but comes with an expensive compu
arXiv:2602.08169v2 Announce Type: replace-cross Abstract: Inference-time steering offers a promising way to control language models (LMs) without retraining. However, standard approaches typically rel
arXiv:2603.10935v4 Announce Type: replace-cross Abstract: Variational autoencoders (VAEs) frequently suffer from posterior collapse, where the latent variables become uninformative as the approximate
arXiv:2605.18636v1 Announce Type: new Abstract: Long-horizon multimodal agents in open-world games must stay goal-directed across many low-level interactions under tight token and latency budgets. Exi
arXiv:2605.18003v1 Announce Type: cross Abstract: Deploying adaptive intelligence at the edge remains challenging due to the high computational and energy cost of training neural models. Spiking Neura
Spotify's Chief Architect just showed how they ship 4,5K deployments /day with Claude at Anthropic stage 27-minutes. free. By #1 music app dev 'More than 99% of our engineers use AI coding tools. Adop
arXiv:2605.18267v1 Announce Type: new Abstract: Normalizing flows (NFs) provide exact likelihoods and deterministic invertible sampling, but have historically lagged behind diffusion models for large-
arXiv:2510.16416v4 Announce Type: replace-cross Abstract: Vision-language models (VLMs) have shown remarkable abilities by integrating large language models with visual inputs. However, they often fai
arXiv:2605.16870v1 Announce Type: new Abstract: Flexible endoscopic robots enable minimally invasive access through natural orifices, but their control accuracy is limited by configuration-dependent h
arXiv:2602.01733v2 Announce Type: replace-cross Abstract: Conformal Prediction (CP) provides a statistical framework for uncertainty quantification that constructs prediction sets with coverage guaran