Retrieval pipelines, vector search, chunking and reranking: how models are grounded in a corpus instead of their weights.
arXiv:2610.28483v1 Announce Type: new Abstract: Learning latent dynamics is fundamental to building action-conditioned world models for planning. Among existing families, Joint Embedding Predictive A...
arXiv:2610.27364v1 Announce Type: cross Abstract: In joint-embedding predictive architectures (JEPA), an encoder and a predictor are trained together to produce a representation that minimizes the er...
arXiv:2610.27515v1 Announce Type: cross Abstract: Pretrained Vision-Language Models (PVLMs) achieve strong zero-shot recognition but often suffer performance degradation under target-domain shift. Un...
arXiv:2610.28160v1 Announce Type: cross Abstract: Programmable Logic Controller (PLC) coding remains a manual, IDE-centric task that demands constant context switching between documentation, project...
arXiv:2610.27079v1 Announce Type: new Abstract: Large language model (LLM) watermarking operates on the next-token distribution that evolves during generation. In streaming generation, each sampled t...
arXiv:2610.27569v1 Announce Type: new Abstract: Tabular foundation models (TFMs) pretrain a transformer on large-scale synthetic priors so that a single forward pass yields zero-shot tabular predicti...
arXiv:2610.27259v1 Announce Type: cross Abstract: The rapid evolution of Large Language Models is fueled by massive datasets, yet this reliance poses severe privacy risks, as models can inadvertently...
arXiv:2603.12976v2 Announce Type: replace Abstract: Scientific discovery increasingly relies on federated learning over data from high-resolution instruments, and these data are strongly non-IID and...
arXiv:2610.26030v2 Announce Type: replace Abstract: Looped transformers reuse a shared block of layers to increase computational depth without increasing the number of trainable parameters. They are...
arXiv:2610.26951v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) is the dominant paradigm for grounding large language models in external corpora, but its similarity-driven, chunk...
arXiv:2610.27241v1 Announce Type: new Abstract: Linear relational embeddings (LREs) approximate a language model's subject-to-object mapping with an affine function. Their faithfulness varies enormou...
arXiv:2610.14531v2 Announce Type: replace Abstract: Chinese historical documents are invaluable cultural treasures, but their complexity poses significant comprehension challenges for non-specialists...
arXiv:2610.27914v1 Announce Type: new Abstract: Real-world observations are noisy and often contain excessive task-irrelevant information. Effective world modeling and planning need to abstract obser...
arXiv:2610.28229v1 Announce Type: cross Abstract: Direct Preference Optimization (DPO) minimizes empirical risk over observed preference comparisons, implicitly assuming that the training preference...
arXiv:2610.28346v1 Announce Type: new Abstract: Autonomous video editing from open-world resources remains a challenging problem in multimedia content creation. Retrieval-based methods are constraine...
DeclineCraft is a dynamic abstention benchmark for retrieval‑augmented generation (RAG) that tests whether models can recognize when neither retrieved documents nor parametric knowledge can answer a question. The benchmark uses unanswerable Minecraft feature‑addition queries that appear after a corpus cut‑off, allowing evaluation of future models. Across 27 large language models, the study measures Correct Abstention Rate (CAR), Answerable Attempt Rate, and Answer Accuracy, finding that single‑shot retrieval boosts CAR, iterative RAG reduces CAR with minimal impact on accuracy, and larger models improve accuracy but not CAR.
MemT-1.0 is a memory-native evolution architecture that embeds memory directly into model computation, using Gated DeltaNet-2 for token-level updates and a Mixture-of-Storage scheme to scale capacity. It is trained in two stages with 34B tokens and released as MemT-1.0-Nano and MemT-1.0-Flash, which outperform parametric memory baselines and compete with retrieval-augmented methods while preserving general capabilities. At a 128K-token history, MemT-1.0-Nano achieves a 1.86× inference speedup and a 66× reduction in persistent storage compared to a full context KV cache.
TRACE is a new framework that reorganizes 168 tasks for text‑vision embedding models into General, Adaptability, and Challenge categories. Experiments on two models show that adding general and adaptation data boosts Adaptability scores by 7.67 and 7.09 points while keeping General performance largely unchanged, yet Challenge performance stays below 5.0 with minimal gains from more data. The study highlights that data adaptation alone cannot overcome weaknesses in multilingual and visual‑document tasks, indicating a need for methodological advances.
The paper introduces CopyGen, a lightweight copy-constrained decoding plugin for frozen large language models that improves exact span reproduction. CopyGen decomposes copying into three decisions—when to copy, how far to copy, and when to stop—using adaptive logit adjustments during verification to emit the longest valid prefix. Across three LLM backbones and four benchmark tasks, it boosts average exact match by 35% relative without finetuning and further improves by 4% with SFT, while also accelerating generation.
CLOSER is an evidence‑aware active visual verification framework for embodied object‑goal navigation. It fine‑tunes a vision‑language model to label each view of a tracked candidate as target, non‑target, or uncertain, then selects complementary views using an embedding‑based diversity gate and fuses their probabilities with a lightweight network. When uncertainty remains, a diffusion policy directs the agent toward the candidate using only the tracked mask and depth until a decisive judgment is reached.