Knowledge catalogue
arXiv:2604.24003v1 Announce Type: new Abstract: Large language models (LLMs) achieve strong reasoning performance by allocating substantial computation at inference time, often generating long and ver
arXiv:2604.23198v1 Announce Type: new Abstract: Current video moment retrieval excels at action-centric tasks but struggles with narrative content. Models can see extit{what is happening} but fail to
arXiv:2604.24156v1 Announce Type: cross Abstract: Efficient and fair spectrum allocation is a central challenge in 6G networks, where massive connectivity and heterogeneous services continuously compe
arXiv:2604.22757v1 Announce Type: cross Abstract: We introduce StratRAG, an open-source retrieval evaluation dataset for benchmarking Retrieval-Augmented Generation (RAG) systems on multi-hop reasonin
arXiv:2601.07698v2 Announce Type: replace Abstract: As emotional support chatbots have recently gained significant traction across both research and industry, a common evaluation strategy has emerged:
arXiv:2601.20597v2 Announce Type: replace Abstract: Continual Text-to-Video Retrieval (CTVR) is a challenging multimodal continual learning setting, where models must incrementally learn new semantic
arXiv:2604.24380v1 Announce Type: new Abstract: While Large Vision Language Models (LVLMs) demonstrate impressive capabilities, their substantial computational and memory requirements pose deployment
arXiv:2604.23475v1 Announce Type: cross Abstract: We study the organization of channel-level importance in transformer feed-forward networks (FFNs). Using a Fisher-style loss proxy (LP) based on activ
arXiv:2509.14635v2 Announce Type: replace Abstract: Understanding and reasoning about entire software repositories is an essential capability for intelligent software engineering tools. While existing
arXiv:2512.07538v3 Announce Type: replace Abstract: Recognizing semantic differences across documents is crucial for text generation evaluation and content alignment, especially in cross-lingual setti
arXiv:2603.26380v2 Announce Type: replace Abstract: The attention mechanism has been the core component in modern transformer architectures. However, the computation of standard full attention scales
arXiv:2604.24346v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly deployed as evaluators in tasks requiring nuanced image understanding, yet their reliability in scoring
arXiv:2604.23806v1 Announce Type: cross Abstract: The reverse process in score-based diffusion models is formally equivalent to overdamped Langevin dynamics in a time-dependent energy landscape. In ou
arXiv:2509.25346v2 Announce Type: replace Abstract: Predicting cellular responses to genetic perturbations represents a fundamental challenge in systems biology, critical for advancing therapeutic dis
arXiv:2604.24088v1 Announce Type: cross Abstract: Handling communication overhead in large-scale tensor-parallel training remains a critical challenge due to the dense, near-zero distributions of inte
arXiv:2509.00072v3 Announce Type: replace Abstract: Post-cutoff performance decay has been widely interpreted as a temporal signal for benchmark contamination. We critically examine this belief and de
arXiv:2604.22880v1 Announce Type: new Abstract: Existing document OCR largely targets plain text or Markdown, discarding the structural and executable properties that make LaTeX essential for scientif
arXiv:2604.24459v1 Announce Type: new Abstract: Despite recent advances in text-to-image generation, models still struggle to accurately render prompt-specified text with correct spatial layout -- esp
arXiv:2604.24155v1 Announce Type: cross Abstract: The quest to align machine behavior with human values raises fundamental questions about the moral frameworks that should govern AI decision-making. M
The California Coastal Commission has issued a formal apology to @elonmusk and SpaceX, adding that it will not consider political views or speech in future regulatory decisions. • The Commission admit
The cracks inside OpenAI are deepening, and the numbers don’t lie. When your own CFO is sounding the alarm, something is seriously wrong. Check this out: 1: OpenAI missed its target of 1 billion weekl
Lily Hay Newman / Wired: The FIDO Alliance launches two working groups to establish industry standards for securing AI agent transactions; Google contributes the Agent Payments Protocol — AI agents ma
arXiv:2604.24083v1 Announce Type: new Abstract: This study introduces the Kerimov-Alekberli model, a novel information-geometric framework that redefines AI safety by formally linking non-equilibrium
arXiv:2604.24749v1 Announce Type: new Abstract: While the optimal sample complexity of binary classification in terms of the VC dimension is well-established, determining the optimal sample complexity
arXiv:2604.23750v1 Announce Type: cross Abstract: Hypernetwork-based methods such as Doc-to-LoRA internalize a document into an LLM's weights in a single forward pass, but they fail systematically on
arXiv:2604.24079v1 Announce Type: cross Abstract: Large Language Models (LLMs) reveal inherent and distinctive personas through dialogue. However, most existing persona discovery approaches rely on su
arXiv:2604.24668v1 Announce Type: new Abstract: Given the increased use of LLMs in financial systems today, it becomes important to evaluate the safety and robustness of such systems. One failure mode
arXiv:2604.22771v1 Announce Type: cross Abstract: Language models cannot be random. This paper introduces Entropic Deviation (ED), the normalised KL divergence between a model's token distribution and
arXiv:2601.15485v2 Announce Type: replace-cross Abstract: Federal research funding shapes the direction, diversity, and impact of the US scientific enterprise. Large language models (LLMs) are rapidly
arXiv:2604.23058v1 Announce Type: cross Abstract: Firms are deploying more capable AI systems, but organizational controls often have not kept pace. These systems can generate greater productivity gai
arXiv:2508.09603v2 Announce Type: replace Abstract: Membership inference attacks serves as useful tool for fair use of language models, such as detecting potential copyright infringement and auditing
arXiv:2511.08577v2 Announce Type: replace-cross Abstract: Improving reasoning abilities of Large Language Models (LLMs), especially under parameter constraints, is crucial for real-world applications.
this feels like less a trend and just something we expect... but q1 had a lot of launches that overlapped with recent startups i've seen: - codex desktop for running parallel - managed agents by claud
This is great - @deepseek_ai V4 supports prefill! :D Most other providers have been dropping support for this critically important capability, so wonderful to see at least one company stepping up. htt
Today we’re releasing Laguna XS.2, Poolside’s first open-weight model. It’s a 33B total / 3B active MoE model built for agentic coding and long-horizon tasks. Trained fully in-house on our own stack.
Together AI announced immediate availability of NVIDIA's Nemotron 3 Nano Omni model to developers through its platform on the day of its release. The Nemotron 3 Nano Omni is a lightweight multimodal m
arXiv:2604.01644v2 Announce Type: replace Abstract: Natural language provides an intuitive way to express spatial intent in geospatial applications. While existing localization methods often rely on d
arXiv:2604.24119v1 Announce Type: new Abstract: Topology reasoning is crucial for autonomous driving. Current methods primarily focus on instance-level learning for centerline detection, followed by a
arXiv:2604.23231v1 Announce Type: cross Abstract: Semantic Communication (SC) backdoor attacks aim to utilize triggers to manipulate the system into producing predetermined outputs via backdoored shar
arXiv:2509.26158v2 Announce Type: replace-cross Abstract: The performance of deep neural networks is strongly influenced by the quality of their training data. However, mitigating dataset bias by manu
arXiv:2603.25158v4 Announce Type: replace Abstract: Equipping Large Language Model (LLM) agents with domain-specific skills is critical for tackling complex tasks. Yet, manual authoring creates a seve
NVIDIA Nemotron-3 Nano Omni is now available to try through Together AI's platform, offering access to a compact multimodal model capable of processing both text and audio inputs. This announcement hi
arXiv:2604.23609v1 Announce Type: new Abstract: Contact-rich manipulation is central to many everyday human activities, requiring continuous adaptation to contact uncertainty and external disturbances
arXiv:2604.23460v1 Announce Type: new Abstract: Chain-of-Thought (CoT) reasoning has emerged as a key technique for eliciting complex reasoning in Large Language Models (LLMs). Although interpretable,
arXiv:2604.24525v1 Announce Type: cross Abstract: Automated code review (ACR) bots are increasingly used in industrial software development to assist developers during pull request (PR) review. As ado
arXiv:2604.23145v1 Announce Type: cross Abstract: Video Question Answering (VideoQA) demands models that jointly reason over spatial, temporal, and linguistic cues. However, the task's inherent comple
Carl Franzen / VentureBeat: US startup Poolside debuts its first open-weight model, Laguna XS.2, a 33B-A3B-parameter MoE model, and Laguna M.1, a proprietary 225B-A23B-parameter MoE model — The AI rac
arXiv:2509.14837v2 Announce Type: replace Abstract: Recent advances in causal interpretability have extended from language models to vision-language models (VLMs), seeking to reveal their internal mec
arXiv:2510.08618v2 Announce Type: replace-cross Abstract: Omni-modal large language models (OLLMs) offer a promising end-to-end solution for slide-enhanced speech recognition due to their inherent mul
Want to see which frontier models do the best on document understanding? Check out our ParseBench leaderboard on @kaggle! https://www.kaggle.com/benchmarks/llamaindex-org/parsebench For more details o
arXiv:2604.22834v1 Announce Type: new Abstract: This paper presents webmcu-vision-web, a single-file, zero-install browser application for end-to-end TinyML vision model training and deployment on the
Welcome to the agentic era! Last week, leaders from our public sector customer and partner ecosystem took the stage at Google Cloud Next to share how they are leveraging AI and agents to scale their i
we're doing a lot more of this, hunting down some of the most annoying bugs in Claude Code let me know if you have any white whales In the last four Claude Code CLI releases, we’ve shipped 50+ stabili
arXiv:2604.23942v1 Announce Type: cross Abstract: People increasingly turn to large language models (LLMs) to interpret ambiguous social situations: a delayed text reply, an unusually cold supervisor,
This post is a question about search provider preferences for use with various open-source tools and APIs (openclaw, pi, opencode, etc.), with examples of options like Brave, SerpAPI, and Gemini. The
arXiv:2604.23398v1 Announce Type: new Abstract: We report a reproducible error pattern in GPT-5.4 on OWL~2~DL compliance queries: the model frequently answers ``unknown'' when the reasoner-entailed an
arXiv:2604.23434v1 Announce Type: cross Abstract: Dynamic Tanh (DyT) removes LayerNorm by bounding activations with a learned tanh(alpha x). We show that this bounding is a regime-dependent implicit r
arXiv:2604.22774v1 Announce Type: cross Abstract: Accurate transcription of handwritten mathematics is crucial for educational AI systems, yet current benchmarks fail to evaluate this capability prope
arXiv:2604.23767v1 Announce Type: new Abstract: Deploying machine learning models across diverse well portfolios requires generalisation to wells with design parameters outside the training distributi
This is a Wordle game result posted by Anthropic on X (formerly Twitter), showing the solution was found in 5 out of 6 possible attempts. The emoji grid indicates the letter placement feedback for eac