Search: “routing”

arXiv:2601.19060v2 Announce Type: replace Abstract: Multimodal Retrieval-Augmented Generation (MM-RAG) augments visual question answering (VQA) with external knowledge, but many systems rely on modula

arXiv:2610.15548v1 Announce Type: new Abstract: Quadruped robots deployed for inspection and field missions must cross heterogeneous terrain, from stairs, slopes, and gaps to high platforms, stepping

arXiv:2610.13982v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models have emerged as an effective paradigm for scaling large language models, yet their massive parameter footprint poses sev

arXiv:2610.16390v1 Announce Type: cross Abstract: A central goal of quantum machine learning is to identify applications in which small quantum circuits can improve task performance while remaining ex

arXiv:2610.18688v1 Announce Type: new Abstract: Agentic visual generation produces experience that can improve the native understanding (agent) and generation roles of a unified multimodal backbone, b

arXiv:2610.16500v1 Announce Type: new Abstract: Image-space reward requires decoding and a separate judge; unified visual generators can instead reuse their understanding pathway for decode-free, self

arXiv:2609.34162v1 Announce Type: cross Abstract: Host-resident layer streaming enables full-parameter LLM fine-tuning beyond GPU memory, but data-parallel ranks compete for shared host resources. Rep

arXiv:2610.15745v1 Announce Type: new Abstract: Multimodal 3D occupancy prediction underpins safe autonomous driving through comprehensive scene understanding. However, existing methods have largely p

arXiv:2610.17065v1 Announce Type: new Abstract: Real-world tables are often heterogeneous, combining numerical features with free-form string columns. Table foundation models (TFMs) excel on numerical

arXiv:2610.16826v1 Announce Type: cross Abstract: Deepfake speech detectors are typically evaluated as binary classifiers on fixed benchmarks, but deployed systems face evolving attacks, codecs, speak

arXiv:2610.14260v1 Announce Type: new Abstract: Augmentation-based self-supervised learning can recover the content shared across views without separating its underlying factors into individual coordi

arXiv:2610.14007v1 Announce Type: new Abstract: A familiar line of reasoning runs as follows: language models are flexible approximators trained under a strictly proper loss, so scale brings them clos

arXiv:2606.18097v2 Announce Type: replace Abstract: Deformable Linear Objects (DLOs), such as wires and cables, are central to industrial assembly. Unlike rigid objects, whose state is captured by a 6

pixel-curator — curation tooling for similarity training (identity, subject, and style models) Repo: https://github.com/snoozey-mf-bnui/pixel-curator What it is pixel-curator is a local-first CLI that

Gguf's: https://huggingface.co/LiquidAI/LFM2.5-24B-A2B-GGUF Free Demo: https://playground.liquid.ai 'LFM2.5-24B-A2B is a 24B-parameter reasoning model with only 2B active parameters, built for fast, p

router lives in the harness, not a gateway, bc the harness knows the task blog: https://www.langchain.com/blog/how-to-build-a-model-router-in-the-harness https://x.com/neil_xbt/status/2109979966859510

New version of my MiniMax H3 timeline workflow released Some very important bug fixes included, and a lot of improvements. The YouTube video above includes a lot of tips for using the workflow in addi

arXiv:2603.03753v2 Announce Type: replace-cross Abstract: The ongoing shift of Artificial Intelligence (AI) from centralized cloud application programming interfaces (APIs) to local inference on edge

arXiv:2610.12927v1 Announce Type: new Abstract: Large language models (LLMs) excel across diverse tasks, benefiting from knowledge priors, in-context reasoning, and test-time computation. Despite thes

arXiv:2610.13570v1 Announce Type: new Abstract: Reasoning models can benefit from test-time scaling, but continued reasoning may become ineffective or harmful, a phenomenon termed overthinking. Evalua

arXiv:2502.11864v2 Announce Type: replace-cross Abstract: Agents in real-world scenarios like automated driving deal with uncertainty in their environment, in particular due to perceptual uncertainty.

arXiv:2610.13328v1 Announce Type: new Abstract: Serving a Mixture-of-Experts (MoE) model pays for expert precision twice: the resident footprint pays for every expert the router can reach, while each

arXiv:2610.13337v1 Announce Type: cross Abstract: Attention sinks are tokens or positions that consistently receive disproportionate attention mass, often beyond what their semantic content would sugg

arXiv:2610.12762v1 Announce Type: new Abstract: Hardware datasheets contain engineering specifications essential for component selection, system integration, and hardware compatibility verification. H

arXiv:2610.13208v1 Announce Type: new Abstract: Time-series foundation models use the same network depth for every input, although forecasting windows differ in structure and difficulty. We test wheth

arXiv:2610.12738v1 Announce Type: cross Abstract: Forecasting rare, high-cost events in multivariate time-series systems is hard because their precursors are seldom observed, so a purely data-driven m

arXiv:2610.12933v1 Announce Type: new Abstract: Large language models have achieved remarkable advances in complex reasoning. However, scaling reasoning post-training with ever-larger corpora can subs

arXiv:2610.13409v1 Announce Type: new Abstract: End-to-end planning in complex scenes must model diverse driving intentions and select the most appropriate trajectory using scene context and future ri

arXiv:2610.13583v1 Announce Type: new Abstract: Detectors of machine-generated text score almost perfectly on scientific-literature benchmarks, implying we can reliably distinguish with these detector

arXiv:2610.12842v1 Announce Type: new Abstract: Neural networks routinely develop representations in which individual neurons contribute to multiple distinct functions, leading to the widespread view

arXiv:2610.13871v1 Announce Type: new Abstract: Many optimization problems provide direct access to only part of the gradient, while the remaining component must be inferred from function evaluations.

arXiv:2605.06212v2 Announce Type: replace Abstract: Attribution methods explain which input features drive a model's prediction, making them central to model debugging and mechanistic interpretability

arXiv:2610.13008v1 Announce Type: new Abstract: In repeated human-AI interaction, a single observation can be quoted, summarized, and returned to a model as multiple reports, making repetition look li

EXL3-QLoRA I've built this over the last few months and I think it's stable enough to share now. A fork of ExLlamaV3, this is a native ExLlamaV3 training implementation with plain PyTorch, without tra

arXiv:2610.13453v1 Announce Type: new Abstract: Reinforcement Learning (RL) substantially improves planning performance in autonomous-driving Vision-Language-Action (VLA) models. However, existing RL

arXiv:2610.12822v1 Announce Type: new Abstract: Multi-agent systems are increasingly used to solve coding tasks by coordinating agents that share code and plans. These agents are often optimized with

arXiv:2610.12670v1 Announce Type: new Abstract: We consider planning for unmanned aerial vehicle (UAV) and unmanned ground vehicle (UGV) teams in aerial monitoring missions with ordering requirements

I've been tuning llama.cpp on my Strix Halo box (Radeon 8060S, 128 GB) and the numbers are good enough to share. It's a branch called rdna35-perf on top of halo-box/strix-llama.cpp, a llama.cpp fork a

arXiv:2610.12939v1 Announce Type: cross Abstract: Decoder hidden states from different layers of large language models (LLMs) often encode complementary information. However, existing speculative deco

arXiv:2610.12651v1 Announce Type: new Abstract: LLM agents executing multi-step workflows fail in two fundamentally different ways: recoverable slips, which further attempts can repair, and knowledge

arXiv:2610.12877v1 Announce Type: new Abstract: Agile and reliable navigation is a core capability for embodied robots navigating complex spatial constraints and dynamic interactions. However, a singl

arXiv:2610.12981v1 Announce Type: cross Abstract: Visual interventions provide a training-free way to reduce hallucinations in multimodal large language models. However, an intervention may have littl

If you build on open-weight reasoning models, you have probably wanted a 'how sure is the model about this answer?' signal that does not cost 5x sampling, a judge model, or a labeled dataset. Hence, w

arXiv:2610.13024v1 Announce Type: cross Abstract: World Action Models (WAMs) inherit powerful priors from pretrained video generation models, but their variational autoencoder (VAE) representations ar

rpc : turn GGML_RPC_DEBUG into a verbosity level and add logs GGML_RPC_DEBUG is now parsed as a number: 0/unset disables debug logs, 1-3 emit increasingly detailed output (events, per-command trace, t

New paper from the team at Sakana AI: Recursive Self-Improvement through Multi-Agent Self-Supervision (MASS) How do we keep improving AI when the tasks become too complex for humans to reliably evalua