Retrieval pipelines, vector search, chunking and reranking: how models are grounded in a corpus instead of their weights.
arXiv:2601.09879v2 Announce Type: replace-cross Abstract: Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report gen...
arXiv:2602.20294v2 Announce Type: replace-cross Abstract: Simulating real personalities with large language models requires grounding generation in authentic personal data. Existing evaluation approa...
arXiv:2609.38932v1 Announce Type: new Abstract: Surface electromyography (sEMG) provides a wearable, camera-free signal for continuous hand-motion inference. Mapping muscle activity to joint kinemati...
arXiv:2609.39547v1 Announce Type: new Abstract: GUI agents built on large language and vision-language models still struggle on unseen applications and complex multi-step tasks, as completing real GU...
arXiv:2609.40335v1 Announce Type: new Abstract: Differentially Private Stochastic Gradient Descent (DP-SGD) is a leading approach for privacy-preserving fine-tuning of large language models (LLMs). M...
arXiv:2609.38312v1 Announce Type: cross Abstract: The Platonic Representation Hypothesis predicts that sufficiently scaled foundation models converge on a shared representation of the world. As each...
arXiv:2609.38545v1 Announce Type: cross Abstract: Machine-learning signals built from financial text treat what institutions say, and what the media repeat, as evidence about value. But whoever shape...
arXiv:2609.39162v1 Announce Type: cross Abstract: Speaker diarization systems based on speaker embeddings and neural diarization exploit complementary forms of speaker information, but their intermed...
arXiv:2508.12145v5 Announce Type: replace Abstract: Recently, autoencoders (AEs) have gained interest for creating parametric and invertible projections of multidimensional data. Parametric projectio...
arXiv:2601.22028v2 Announce Type: replace Abstract: Most LLM unlearning methods aim to approximate retrain-from-scratch behaviors with minimal distribution shift, often via alignment-style objectives...
arXiv:2605.07938v2 Announce Type: replace Abstract: Single-cell representation learning (SCRL) from gene expression data offers a way to uncover the complex regulatory logic underlying cellular funct...
arXiv:2609.19011v2 Announce Type: replace Abstract: Knowledge distillation can copy a deployed model by training a student on its logits or features. The student inherits a watermark only through the...
arXiv:2610.01385v1 Announce Type: cross Abstract: This work aims to answer the question of whether it is possible to generate irreversible protected templates when the PolyProtect biometric template...
Build2SPARQL is a large-scale benchmark dataset for translating natural-language questions into SPARQL queries over building knowledge graphs. The dataset is generated by a KG‑grounded pipeline that produces 6,136 executable SPARQL queries and 30,680 corresponding natural-language questions across six query-pattern families and five vocabulary registers, covering 201 building KGs. Human validation shows high semantic fidelity, naturalness, and operational plausibility, and retrieval‑augmented evaluation demonstrates significant accuracy gains for open‑weight language models.
NextMe-800 is an approximately 800‑hour first‑person video dataset collected from a single volunteer over 126 days, featuring 1 Hz images, gaze, and audio. The data are captioned at five hierarchical abstraction levels—from atomic actions to major activities—enabling personalized action anticipation as an open‑vocabulary K‑step sequence prediction task. The authors also introduce NextAct, a 1,500‑point benchmark that combines NextMe‑800 with the multi‑person EgoLife dataset, and evaluate models using an embedding‑based soft edit distance to assess how well personal behavior can be anticipated across abstraction levels and prediction horizons.
ScholarCatalyst is a new benchmark that evaluates how well AI systems can retrieve research papers that inspire new work. The dataset was created by having 184 lead authors of 207 recent computer science papers annotate which earlier papers helped their projects, providing detailed rationales. The benchmark tests retrieval from the literature available at the start of a project, revealing that current agentic search and even advanced models like Claude Fable 5.1 perform only modestly better than simple embedding retrieval.
arXiv:2610.00604v1 Announce Type: cross Abstract: Vision-language-action policies often see only one or a few recent frames, which makes it difficult to evaluate how they use information that disappe...
arXiv:2610.00676v1 Announce Type: cross Abstract: Unsupervised skill discovery has emerged as a promising approach for leveraging reward-free datasets to pretrain general-purpose policies. However, c...
arXiv:2610.00006v1 Announce Type: new Abstract: Pretrained Vision Transformers encode whether two image patches belong to the same object. This IsSameObject signal is decodable from frozen patch embe...
arXiv:2610.00571v1 Announce Type: cross Abstract: Large reasoning models (LRMs) have achieved substantial improvements in solving complex mathematical problems, but often produce lengthy, repetitive,...