Recurrent architectures that compress history into a single state are cheap but have classically been unable to exactly retrieve far past key–value bindings, whereas attention content-addresses the whole past at quadratic cost. We show that the conflict between dense language modeling and long-range associative recall...
Ziheng Zhou· Zenodo (CERN European Organi...· 0 citations
Here we present four companion documents for the program corpus listed below. The Taxonomy of Claims lists the theory's claims and assigns each one a status. The claims are grouped in four tiers: Framework (the laws and constants, which the theory does not derive), Derived from Framework (quantities computed from the f...
Scott Weller· Zenodo (CERN European Organi...· 0 citations
Abstract Indigenous and nomadic communities worldwide face disproportionate and persistent barriers to reproductive health services, including family planning and safe abortion. The Raute of Nepal — one of the country’s last nomadic hunter-gatherer groups represent a uniquely marginalized population for whom no prior p...
Manoj Joshi, Anjali Bhatt, Sudip Khanal et al.· PLOS Global Public Health· 0 citations
VisCo is a training-efficient self-compression framework that reuses the pretrained VLM itself as an intrinsic compressor that compresses visual information using a small set of memory tokens and transfers hierarchical information from encoding to decoding.
Yupeng Zheng, Kai Zou, Bin Liu et al.· arXiv.org· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A sum-token-goodput maximization problem that jointly accounts for mode selection, draft-length control, and power allocation is formulated, and a simple optimal structure is revealed that enables efficient search over the number of UL devices, with the corresponding transmit powers optimized accordingly.
HSMLog, a two-stage framework for HSM log anomaly detection with retrieval-grounded behavioral analysis, is presented, demonstrating effective anomaly alerting and incident triage in the studied setting.
This work extends the concept of gadget game, tabular technique for test-time search, to the reinforcement learning setting and formally proves that, unlike prior tabular algorithms, regularized policy-gradient algorithms limit possible strategy degradation caused by test-time reasoning, even without the gadget games.
Ondrej Kubícek, Viliam Lisý, Tuomas Sandholm· 1 citation
VisLens (Visual Focus via Logit Lens), a Visual Search method built on the logit lens, which decodes the semantics held in a hidden state by projecting it through the LLM head, and which matches or exceeds prior baselines while delivering a substantial latency advantage.
Jingfeng He, Sanghwan Kim, Zeynep Akata· 0 citations
FAIRLENS shows that fair high-stakes VLM behavior requires similar treatment across groups and refusal to infer high-stakes attributes from appearance, and its question suite transfers to any face corpus with demographic annotations.
Vahid Reza Khazaie, Ahmed Y. Radwan, Shaina Raza· 0 citations
This work studies why pretrained language models often fail to preserve mixtures of many components and shows that exact preservation generally requires context-dependent correction, whose required dimensionality can grow with the number of components.
Declarative Attention is introduced, a protocol that elicits the model to declare where it needs to attend within its chain-of-thought, partitioning generation into three modes:(full context),(a specific region), and(recent output only).
Namgyu Ho, Huzama Ahmad, Woosung Koh et al.· 1 citation
DeepAffinity is proposed, which leverages Small Language Models with structured prompts and specialized prediction heads fine-tuned for this task, and shows DeepAffinity outperforms standard generative fine-tuning methods, while general-purpose open-source LLMs perform poorly without task-specific tuning.
Yotam Eshel, Guy Hadad, Guy Feigenblat et al.· 0 citations
