Knowledge catalogue
arXiv:2609.35810v1 Announce Type: cross Abstract: Large language models are increasingly used in oncology applications, but their predictions are often weakly grounded in explicit medical structure. W
NVIDIA NeMo Relay supplies a common observability layer that captures ordered lifecycle events and structured trajectories—ATOF event streams, ATIF step‑by‑step footprints, and OpenTelemetry spans wit
arXiv:2609.35822v1 Announce Type: new Abstract: Sycophantic agreement in language models refers to the tendency to overly affirm a user's stated beliefs or preferences, often at the expense of factual
arXiv:2609.37656v1 Announce Type: new Abstract: Large vision-language models (LVLMs) exhibit strong reasoning capabilities, yet the visual and textual evidence supporting the generated responses remai
arXiv:2609.35674v2 Announce Type: replace Abstract: Of the approximately 4,500 Oracle Bone Inscription (OBI) characters discovered from the Shang dynasty, only about 1,600 have been deciphered. Many c
arXiv:2609.36924v1 Announce Type: cross Abstract: Learning humanoid skills from videos typically requires a successful human demonstration, which often demands custom data collection. Although failure
arXiv:2609.37132v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) improves large language models by letting a self-teacher with privileged information provide dense token-level superv
arXiv:2512.18991v3 Announce Type: replace-cross Abstract: Dominant paradigms for 4D LiDAR panoptic segmentation are usually required to train deep neural networks with large superimposed point clouds
arXiv:2609.36316v1 Announce Type: cross Abstract: Evaluation awareness (EA) can cause large language models (LLMs) to behave differently during audits than in deployment, yet measuring and accounting
arXiv:2609.36530v1 Announce Type: new Abstract: Motion planning often admits multiple feasible solutions, making multimodal generation valuable, particularly for flexible multi-robot coordination. Dif
arXiv:2609.37169v1 Announce Type: cross Abstract: Mid-training equips pretrained large language models with specialized and reasoning capabilities, but the returns of this stage are bounded since addi
arXiv:2609.36585v1 Announce Type: new Abstract: Pretrained transformers use little of their depth to follow references in context. Thirteen base models reliably follow only 1.4-3.6 lines, and extra pr
arXiv:2609.37279v1 Announce Type: new Abstract: Neural solvers offer efficient surrogates for numerical simulation of partial differential equations (PDEs). For time-dependent problems, strong one-ste
arXiv:2609.36237v1 Announce Type: new Abstract: Many learning tasks require stability to small transformations while retaining sensitivity to larger ones. We introduce transversal pooling neural netwo
arXiv:2609.37082v1 Announce Type: new Abstract: Long-horizon information-seeking agents often accumulate noisy or misleading context, causing early mistakes to persist and making recovery increasingly
arXiv:2609.38081v1 Announce Type: new Abstract: On a single task, deep networks can learn many solutions, depending on their optimizer, training data, architecture, and hyperparameters. Many of these
arXiv:2609.37581v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) excel at visual understanding and reasoning but often incur substantial inference costs due to the large number of visua
arXiv:2609.36529v1 Announce Type: cross Abstract: Recurrent neural networks (RNNs) compress the historical context into a memory state of fixed size, thus allowing for constant-time inference. The mem
arXiv:2609.37241v1 Announce Type: new Abstract: User-written Triton kernels enable high-performance GPU computation within PyTorch, but their end-to-end latency can remain dominated by host-side orche
arXiv:2609.36372v1 Announce Type: cross Abstract: Distortion-free watermarking enables reliable attribution of machine-generated text while preserving output distribution. However, existing methods op
We had a Dell B300 in the lab for a few weeks and used it to create two fine tunes of Qwen Flash Next. Victoria (coding and agents) Qwen3.8-Flash-Next cut down by 44% using a paper / technique called
arXiv:2609.37339v1 Announce Type: new Abstract: General multi-object tracking (GMOT) tracks all instances of a user-specified category from a single first-frame exemplar. Prior work relies on bounding
arXiv:2603.28773v2 Announce Type: replace-cross Abstract: Large language models (LLMs) frequently generate confident yet factually incorrect content when used for language generation (a phenomenon oft
arXiv:2609.36980v1 Announce Type: new Abstract: Despite recent advances in accuracy and efficiency, coarse matching remains an indispensable yet costly stage in existing semi-dense matchers due to den
A LoRA designed to reduce the typical smooth/plastic AI look and bring natural texture and realism to your images. What's new in V5 (KL9_V5_Pro): Summary: This version gives the best natural texture s
arXiv:2609.35913v1 Announce Type: cross Abstract: Code large language models acquire programming capabilities from large code corpora, but can also memorize implementations that later require removal.
arXiv:2609.36063v1 Announce Type: cross Abstract: Neural Combinatorial Optimization (NCO) has achieved strong empirical success, yet the internal mechanisms driving model decisions remain largely unex
arXiv:2609.36262v1 Announce Type: new Abstract: Recent studies have observed that parameter changes during language-model post-training can be concentrated in a small subset of coordinates. This pheno
arXiv:2609.36678v1 Announce Type: new Abstract: Private Evolution (PE) is a differentially private algorithm for synthetic data generation. While it can be viewed as a Wasserstein learning algorithm,
arXiv:2609.37264v1 Announce Type: cross Abstract: Affordance perception aims to localize actionable regions supporting embodied interaction, yet 2D and 3D affordance grounding have evolved as separate
arXiv:2609.37031v1 Announce Type: new Abstract: Building extraction from optical remote sensing (RS) imagery is fundamental to urban mapping, yet existing methods are often dataset-specific and genera
arXiv:2609.37076v1 Announce Type: new Abstract: Large language models trained on vast corpora inherently risk memorizing harmful content that may later re-emerge in their outputs. To mitigate this iss
arXiv:2609.37119v1 Announce Type: cross Abstract: Recent approaches to reinforcement learning (RL) post-training for large language models increasingly remove the critic to reduce training instability
Hi folks, I've spent the last couple of months experimenting with Strix Halo and previously I released a proof of concept I called llama-halo-hybrid. I've continued updating it and it now performs ver
loss curve (taken from the last micro of every step, to explain the variation) some help from gemini 3.8 flash high About 40% of the way done with the initial fine tune. The loss is so spiky because I
arXiv:2609.36805v1 Announce Type: new Abstract: Large language model (LLM) agents reuse external memory to guide new tasks, but effective retrieval requires learning which memory sets improve executio
arXiv:2609.37772v1 Announce Type: new Abstract: Diffusion and flow-matching Vision-Language-Action (VLA) policies generate action chunks through iterative denoising, incurring substantial inference la
arXiv:2609.38043v1 Announce Type: new Abstract: Interactive agent benchmarks and multi-turn reinforcement learning increasingly place a second language model in the role of the user. This simulated us
Apple Foundation Model 3 through Private Cloud Compute is directly available in macOS 27.2 through familiar AI clients. For Mac users running local models such as Qwen 3.x or Gemma 4, AFM 3 PCC is a c
arXiv:2609.37198v1 Announce Type: new Abstract: Pretrained text-to-image models contain broad visual knowledge, yet they cannot reliably acquire or refine a specific visual identity from only a few re
arXiv:2609.37250v1 Announce Type: cross Abstract: World-action models (WAMs) couple future visual-state prediction with action generation. By adapting video generators or image-editing models pretrain
arXiv:2609.37098v1 Announce Type: cross Abstract: Vehicle-infrastructure cooperation can complement onboard sensing with broader and more informative observations of the traffic environment, providing
arXiv:2609.36804v1 Announce Type: cross Abstract: Chinese Semantic Error Correction (CSEC) targets semantic errors in Chinese text, which are typically more subtle and complex than spelling and gramma
arXiv:2609.37105v1 Announce Type: cross Abstract: Language model agents can be improved by updating their model weights or refining the harness that guides task execution. These components are coupled
arXiv:2609.37435v1 Announce Type: new Abstract: Neural Koopman autoencoder models have been shown to successfully build a latent embedding with linear dynamics for arbitrary dynamical systems, enablin
arXiv:2609.36911v1 Announce Type: new Abstract: In this thesis I develop methods for statistical inference when the distributions arising from complex biological systems are multi-modal, geometrically
arXiv:2609.35916v1 Announce Type: cross Abstract: Real-world embodied agents often pursue independent objectives within a shared physical environment, where their actions can alter the conditions face
arXiv:2609.37457v1 Announce Type: new Abstract: Enterprise artificial-intelligence agents increasingly call tools, modify infrastructure, and process protected data, creating a need to separate action
arXiv:2609.37003v1 Announce Type: new Abstract: Vessel perception from space is crucial for a wide range of maritime applications, from traffic monitoring to environmental protection. However, most ex
arXiv:2609.36870v1 Announce Type: new Abstract: Video demonstrations offer a scalable alternative to costly robot data for learning manipulation, yet existing reconstruction-based approaches often rel
arXiv:2609.37950v1 Announce Type: new Abstract: Video understanding agents acquire evidence through an executable harness that controls what they observe and how they use those observations. However,
arXiv:2609.36691v1 Announce Type: new Abstract: Manipulation behaviors vary widely across objects and scenes, but they share a small set of reusable skills, and planning with these skills helps embodi
arXiv:2609.37519v1 Announce Type: cross Abstract: Video-based policy learning is particularly promising, as it illustrates target behaviors without requiring action annotations or embodiment-matched d
arXiv:2609.38119v1 Announce Type: new Abstract: Long-form video understanding requires multimodal agents to iteratively gather evidence over many reasoning steps. However, most existing agentic method
Perplexity AI released a preview of its embedding model “pplx‑embed‑v2‑context‑9b” on Hugging Face on September 30, 2026 (7:07 PM local time). The Hugging Face repository page displays 3 122 views and
arXiv:2609.37709v1 Announce Type: cross Abstract: Recent multimodal image generation models can take multiple images and textual instructions as input, enabling reference-based generation guided not o
arXiv:2609.37287v1 Announce Type: cross Abstract: Image translation is a fundamental capability of multimodal models for multilingual applications, requiring visual understanding and meaning preservat
arXiv:2609.38086v1 Announce Type: new Abstract: Active multimodal agents use visual tools to acquire task-relevant evidence while reasoning. Although reinforcement learning samples multiple interactio
arXiv:2609.37324v1 Announce Type: new Abstract: Conflicting emotional cues can be individually valid: a subdued voice may reflect a blocked goal while a smile satisfies a social obligation. Their inte
arXiv:2609.37360v1 Announce Type: new Abstract: Defect detection systems for industrial condition monitoring can only be relied upon if they are validated, yet defective samples are rare and, for a sp