Knowledge catalogue
arXiv:2605.22884v1 Announce Type: cross Abstract: Autoregressive Transformer KV caches grow linearly with context length; sliding-window caching bounds memory but discards evicted tokens entirely, so
NVIDIA's Pixel Diffusion Decoder (PiD) is an open-source decoder that replaces VAE decoders in image generation pipelines without retraining, producing sharper fine details and textures through a lear
The raw reasoning stream can be accessed through the message.thinking field in the API response or the thinking endpoint field, which contains the reasoning trace separately from the final answer. Use
arXiv:2605.23087v1 Announce Type: new Abstract: Neural collapse (NC) describes the structured geometry that emerges in the features and weights of trained classifiers. Recent theory suggests NC can be
arXiv:2605.23081v1 Announce Type: new Abstract: Efficient attention algorithms are critical to mitigate the quadratic cost of attention in long-context workloads. Prior work utilises block-scaled quan
arXiv:2602.19174v5 Announce Type: replace Abstract: Natural language processing for the Turkic language family, spoken by over 200 million people across Eurasia, remains fragmented, with most language
arXiv:2605.23565v1 Announce Type: cross Abstract: Reinforcement learning agents often exhibit unintended goal-directed behaviour outside their training distribution, but we currently lack a principled
v0.30.0-rc25 is a pre-release version of Ollama that changes the architecture to directly support llama.cpp instead of building on top of GGML, and allows for compatibility with GGUF file format. MLX
arXiv:2605.23717v1 Announce Type: new Abstract: Autonomous landing of Unmanned Aerial Vehicles on maritime vessels is challenging due to the coupled motion of the vehicle and landing platform in open-
arXiv:2605.22876v1 Announce Type: new Abstract: Existing neural solvers for Multi-Objective Combinatorial Optimization Problems (MOCOPs) commonly adopt decomposition-based strategies that scalarize an
arXiv:2605.23719v1 Announce Type: cross Abstract: Vision Transformers have achieved remarkable success in computer vision, but their common use of learnable one-dimensional positional encodings weaken
arXiv:2605.22873v1 Announce Type: cross Abstract: Chain-of-thought (CoT) reasoning has become the default strategy for enhancing LLM capabilities, yet its application raises a fundamental question: wh
ComfyUI_SamplingUtils with Klein_9B enables rapid image style changes within ComfyUI workflows, supporting text-to-image generation and reference-based editing for quick style transformation and conte
v0.30.0-rc24 is a pre-release version that changes Ollama's architecture to directly support llama.cpp instead of building on top of GGML, enabling compatibility with the GGUF file format. MLX is used
Nous Research shared an update on X regarding progress made by their team on research or development initiatives. Without access to the specific post content, this likely documents advances in AI rese
arXiv:2411.01332v5 Announce Type: replace Abstract: Despite significant advancements in XAI, scholars note a persistent lack of solid conceptual foundations and integration with broader scientific dis
arXiv:2605.22432v1 Announce Type: new Abstract: Modern deep learning commonly relies on AdamW with prescribed learning rate schedules, but recent works challenge both components: Schedule-Free optimiz
another of the dudes going after me (Theo) is perhaps paid by OpenAI 🙄 He doesn’t work for them, but he likely has or has had some paid partnership with them, because they’ve invited him onto their ca
arXiv:2605.22372v1 Announce Type: new Abstract: Vision Transformers (ViTs) face severe computational bottlenecks due to the quadratic complexity of self-attention at high resolutions. Existing token r
arXiv:2605.21820v1 Announce Type: new Abstract: Self-driving laboratories or autonomous experimentation are emerging as transformative platforms for accelerating scientific discovery. Bayesian optimiz
arXiv:2605.22703v1 Announce Type: new Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a central paradigm for scaling LLM reasoning, yet its optimization often suffers fr
arXiv:2605.21600v1 Announce Type: new Abstract: Computational antibody CDR design methods condition on antigen structure to generate binding loops, yet existing architectures conflate two fundamentall
arXiv:2509.08933v2 Announce Type: replace Abstract: We study the problem of learning the optimal policy in a discounted, infinite-horizon reinforcement learning (RL) setting in the presence of adversa
arXiv:2605.22379v1 Announce Type: cross Abstract: With the advancement of science and technology, the importance of emotion research has become increasingly evident. Electroencephalography (EEG)-based
arXiv:2605.21542v1 Announce Type: new Abstract: Country-level temporal panels are widely used in empirical analysis. Researchers often need to audit how different entities respond to historical signal
arXiv:2605.21563v1 Announce Type: new Abstract: Recent reviews find that the vast majority of published healthcare federated learning (FL) studies never reach real-world deployment. We developed an em
arXiv:2605.22613v1 Announce Type: new Abstract: Recent LLM-guided evolutionary search methods have shown that iterative program mutation can discover strong algorithms, but they typically optimize eac
arXiv:2605.22779v1 Announce Type: cross Abstract: Production systems generate millions of log lines daily, yet most anomaly detectors operate at the session or window-level, flagging groups of lines r
arXiv:2605.22612v1 Announce Type: cross Abstract: Benchmarks are necessary for healthcare evaluation, but are not sufficient for predicting deployment performance. Our position is that the evaluation-
arXiv:2603.02604v2 Announce Type: replace Abstract: We introduce Heterogeneous Agent Collaborative Reinforcement Learning (HACRL), a new Reinforcement Learning from Verifiable Reward (RLVR) problem th
arXiv:2605.22223v1 Announce Type: new Abstract: We study how we can leverage only a handful of characteristics of a transformer's architecture to closely predict the number of different sequences it c
arXiv:2605.22410v1 Announce Type: new Abstract: Spectral clustering largely depends on the affinity graph, yet constructing a graph that preserves reliable local connectivity while adapting to heterog
arXiv:2605.22794v1 Announce Type: cross Abstract: Autonomous agentic systems are largely static after deployment: they do not learn from user interactions, and recurring failures persist until the nex
arXiv:2605.21519v1 Announce Type: cross Abstract: Graph Partitioning is a critical problem in numerous scientific and engineering domains including social network analysis, VLSI design, and many more.
Patina @patinaglobal is one of the coolest and most seminal @EarthlingVC portcos Even as a fund that's very focused on immersive computing (i.e. physical computing) I've seen very, very few things goi
Peekaboo is likely a technical concept or tool discussed by Jerry Liu, co-founder of LlamaIndex, based on the X platform post. Without access to the specific tweet content, this probably covers an AI
arXiv:2605.21859v1 Announce Type: cross Abstract: Phylogenetic trees are hybrid objects: branch lengths vary continuously, while topologies change discretely through edge contractions and expansions.
arXiv:2605.22746v1 Announce Type: new Abstract: Real-world sensor-based learning systems require uncertainty estimation that is both reliable and computationally efficient. Evidential Deep Learning (E
arXiv:2605.21765v1 Announce Type: new Abstract: The practical adoption of sampling-based inference (SAI) in Bayesian neural networks (BNNs) remains limited, partly due to persistent misconceptions abo
arXiv:2605.22055v1 Announce Type: new Abstract: Time Series Classification (TSC) is a long-standing research problem that has gained increasing attention in recent years with the rapid growth of large
arXiv:2605.21780v1 Announce Type: new Abstract: Randomized smoothing is a powerful tool for certifying robustness to adversarial perturbations, including poisoning attacks via randomized training and
arXiv:2605.22738v1 Announce Type: new Abstract: Shapley and Banzhaf interactions capture the complex dynamics inherent in modern machine learning applications. However, current estimators for these hi
arXiv:2605.22097v1 Announce Type: cross Abstract: Photonic quantum computing is a promising platform for scalable quantum machine learning, but designing effective hybrid architectures remains challen
arXiv:2605.22041v1 Announce Type: cross Abstract: While RAG systems are increasingly deployed in dynamic web search, temporal volatility amplifies their vulnerability to adversarial attacks. Existing
arXiv:2605.22719v1 Announce Type: new Abstract: We report a small, reproducible audit of which sparse-autoencoder (SAE) features of GPT-2 small fire differently on failed versus successful trials of t
arXiv:2602.10894v2 Announce Type: replace Abstract: Two-player games such as board games have long been used as traditional benchmarks for reinforcement learning. This work revisits a policy optimizat
arXiv:2505.22749v2 Announce Type: replace-cross Abstract: Attractor dynamics are a hallmark of many complex systems, including the brain. Understanding how such self-organizing dynamics emerge from fi
arXiv:2605.21724v1 Announce Type: new Abstract: Hyper-Connections (HC) improve residual networks by introducing learnable mixing across multiple residual streams, but unconstrained mixing leads to tra
arXiv:2605.22800v1 Announce Type: new Abstract: Robustness, domain adaptation, photometric and occlusion invariance, compositional generalisation, temporal robustness, alignment safety, and classical
arXiv:2605.22723v1 Announce Type: new Abstract: A central error measure in Gaussian DDPMs is the path-space KL divergence between the exact reverse chain and the learned Gaussian reverse process. This
arXiv:2605.22481v1 Announce Type: new Abstract: Backdoor poisoning attacks behave counter-intuitively in high dimensions: stronger training triggers can help the defender. We study regularised general
arXiv:2605.22342v1 Announce Type: new Abstract: While 4D Gaussian Splatting (4DGS) has revolutionized high-fidelity dynamic reconstruction, safeguarding the intellectual property of these assets remai
arXiv:2404.05307v2 Announce Type: replace Abstract: Reliable people detection is crucial for the safe autonomy of mobile robots and heavy vehicles, both on roads and in industrial settings like mining
arXiv:2605.22216v1 Announce Type: new Abstract: This report presents our solution for the WeatherProof Dataset Challenge, namely CVPR 2026 8th UG2+ Challenge Track 2: Semantic Segmentation in Adverse
arXiv:2605.20456v1 Announce Type: cross Abstract: Agentic AI coding systems can inspect repositories, plan implementation steps, edit files, call tools, run tests, and submit pull requests. These capa
arXiv:2508.03865v4 Announce Type: replace Abstract: Some Question Answering (QA) systems rely on knowledge bases (KBs) to provide accurate answers. Entity Linking (EL) plays a critical role in linking
arXiv:2605.22631v1 Announce Type: new Abstract: Accurately reconstructing full-body poses from sparse head and hand trajectories is a foundational challenge for immersive AR/VR telepresence. Current m
Release b9283 of llama.cpp includes updates to the BPE tokenizer with new pre-type dispatch and Carbon-3B DNA tokenizer support, along with refactoring of tokenizer helpers to follow project conventio
B9286 introduces support for HybridDNATokenizer as its own vocabulary type (LLAMA_VOCAB_TYPE_HYBRIDDNA), enabling proper tokenization of DNA sequences by routing raw text through a DNA-aware splitter
arXiv:2605.22123v1 Announce Type: new Abstract: Designing reward functions that generalize beyond controlled laboratory settings remains a fundamental challenge in reinforcement learning for robotics.