Leaderboards, eval harnesses and ablations — the contested business of deciding which model is actually better.
arXiv:2609.36590v1 Announce Type: cross Abstract: Self-speculative decoding accelerates large language model (LLM) inference by drafting tokens from the target model itself, but faces a sharp tradeof...
arXiv:2601.21225v4 Announce Type: replace-cross Abstract: Large language models have made substantial progress in mathematical reasoning. However, benchmark development for multilingual evaluation ha...
The paper introduces Alignment Forecasting, a method for predicting whether fine‑tuning a language model on a given dataset will increase specific alignment failures such as deception or sycophancy. It presents ALIGNMENTFORECASTBENCH, a benchmark of over 5,000 forecasting questions across many models, datasets, and failure modes, and shows that a simple forecasting scaffold using an LLM’s assessment of dataset bias can outperform baseline forecasters. The authors demonstrate that filtering out high‑risk training examples identified by the forecaster can improve alignment in multiple‑choice evaluations, though benefits in open‑ended conversations remain uncertain.
arXiv:2609.37834v1 Announce Type: new Abstract: Harness optimization provides a practical setting for recursive self-improvement (RSI), where agent-generated modifications inform subsequent changes t...
arXiv:2609.36969v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) is a state-of-the-art technique for 3D scene rendering, offering high efficiency and excellent visual quality. However, be...
arXiv:2609.37400v1 Announce Type: new Abstract: Generating realistic 3D dance from music is a challenging task that requires accurate synchronization with musical rhythms while capturing the spatial...
The study investigates whether large language models (LLMs) exhibit reward valuation mechanisms analogous to human anhedonia by applying clinical tests designed for major depressive disorder. Researchers identified reward‑anticipatory units in state‑of‑the‑art AI models, showed that perturbing these units predicts Nucleus Accumbens activity, and caused the models to choose low‑effort, low‑reward tasks—mirroring human anhedonia. The findings suggest that specific reward‑valuation circuits in AI can functionally resemble those in humans, providing a mechanistic bridge between computational and neurobiological models of motivation.
arXiv:2609.36760v1 Announce Type: new Abstract: Multi-Head Latent Attention (MLA) enables expressive multi-head attention with compact caches for its content and decoupled RoPE paths, yet cache memor...
arXiv:2609.37888v1 Announce Type: cross Abstract: Class-Incremental Learning (CIL) requires models to recognize new classes over time without forgetting previously learned ones. With the rise of visi...
arXiv:2609.37944v1 Announce Type: cross Abstract: A wide range of methods have been proposed, including physics-informed neural networks, which are powerful but do not guarantee identifiability of th...
arXiv:2609. 37968v1 Announce Type: new Abstract: Advances in the coding capabilities of LLM agents allow them to inspect and modify their own instructions, tools, and execution procedures.
arXiv:2609.37574v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used to enrich user queries in information retrieval (IR) so that a standard retriever such as BM25 can...
ByteTraX is a lightweight enhancement to the ByteTrack multi‑object tracking architecture that introduces a single unified matching threshold and stricter track initiation criteria to reduce erroneous track reclassification and identity switches. The method yields consistent performance gains across several benchmarks—GMOT‑40, LC‑MOT, SportsMOT, TeamTrack, DAMUNT, and DeepSea‑MOT—while boosting processing speed by over 10%. Quantitatively, ByteTraX achieves more than a 40% drop in identity switches, with mean improvements of 3.6 in HOTA, 5.6 in IDF1, and 6.3 FPS.
arXiv:2609.35868v1 Announce Type: new Abstract: Is human readability necessary for effective fine-tuning of large language models? We investigate whether model-conditioned training representations ca...
arXiv:2609.37969v1 Announce Type: new Abstract: High-resolution video generation is expensive, as its cost grows rapidly with the number of spatiotemporal tokens. A practical alternative first genera...
arXiv:2609.36601v1 Announce Type: new Abstract: On-policy distillation (OPD) reduces train-test state mismatch by training a student on its own generated trajectories, but weak students may visit tea...
AnyAct introduces a universal action layer that consolidates diverse tool capabilities into a self‑evolving action space for AI agents operating in open‑world environments. It tackles the scale dilemma, tool non‑stationarity, and heterogeneous feedback by using hierarchical progressive retrieval and test‑time reliability evolution, while a heterogeneous observation grounding module unifies multi‑modal feedback. Evaluations on LiveMCPBench and the newly created OSMCP benchmark show state‑of‑the‑art performance, with significant gains in task success rate and reduced execution steps, especially for models with limited native capabilities.
arXiv:2609.36934v1 Announce Type: new Abstract: Traffic signal control (TSC) is essential for improving urban mobility and reducing congestion. Although roadside cameras are widely deployed at signal...
THEIA is a multimodal dataset that pairs thousands of analog circuit layout images with question‑answer conversations, and it introduces a benchmark using a fine‑tuned vision‑language model to analyze GDSII files. The dataset and benchmark enable designers to interact with and query physical layouts as intuitive, meaningful entities. Experiments on five realistic tasks show the fine‑tuned model outperforms general‑purpose vision‑language models by up to 73%, revealing a significant gap between general multimodal reasoning and domain‑specific layout understanding.
NeuronSifter is a framework for planning interventions in central nervous system microenvironments by converting treatment regimens into state‑conditional target‑occupancy fields and propagating them through microenvironment dynamics. It selects measurements based on their expected reduction in intervention loss, integrating typed outcomes into a unified posterior. In synthetic Alzheimer’s disease scenarios, occupancy conditioning improves trajectory probability scores and intervention ordering accuracy, and decision‑directed acquisition reduces terminal risk compared to a Bayesian experimental design planner.