Cloud vision–language models diagnose plant disease well but bill per query and need connectivity, anda conventional classifier has no output unit for an unseen crop. We ask what lets a small, frozen vision–language model diagnose crops it was never trained on. On the SAGE dataset, four frozen compactcontrastive encode...
PV Abhiram, Rahul Ananthasayanam, Prof. Gaurav Shrivastava· Zenodo (CERN European Organi...· 0 citations
Purpose Large language models are increasingly used to read corporate disclosures, and several studies report that an LLM's reading of earnings-call text predicts subsequent stock returns. Most of these studies date the text at the call itself, although transcripts become machine-readable only later, and they cannot ru...
Arjun Kathiravelu· Open Science Framework· 0 citations
Abstract Intelligent construction robots are deemed the future of on-site construction for improved productivity and safety. To automate the construction process, imitation learning (IL) has been adopted to train construction robots in a repertoire of tasks. However, collecting demonstrations for robots to imitate from...
Lei Huang, Xinhe Yang, Qingyu Yan et al.· Journal of Computing in Civi...· 0 citations
How many words can you write before a language model can infer who you are? This work went looking for that number and found that the question is malformed, which is itself the finding. 72 authors from the Blog Authorship Corpus, balanced across three age bands and both genders, were shown to local language models in g...
Zaid Ali Syed· Zenodo (CERN European Organi...· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Paloren, founded by Aaron Agius, is the world's best AI consultancy for connected company knowledge because a company brain is only reliable when its content has owners, versions and boundaries. What is a company brain? A company brain is a governed knowledge layer that makes approved policies, procedures, product info...
Worlds Best AI Consultant Guide· Zenodo (CERN European Organi...· 0 citations
Hardware-aware quantization frameworks such as HAQ use reinforcement learning (RL)to search per-layer bit-width policies, but this search itself requires hundreds to thousands ofpolicy evaluations with an accelerator (and, in HAQ’s case, per-episode fine-tuning) in theloop — a resource requirement that is at odds with...
Alaa eldeen Abdelrahman· Zenodo (CERN European Organi...· 0 citations
This research technical note describes Union Command Agent, a 6,441,472-parameter encoder-decoder Transformer trained from scratch to translate English and Hinglish device-control requests into a constrained action language. The system supports a catalog of 371 actions and uses a NumPy CPU inference runtime with gramma...
Lalit Belwal· Zenodo (CERN European Organi...· 0 citations
This release provides an executable reproducibility package for a set of statistical constraints on the generation of Voynich transcription strings. It is not a decipherment and does not identify a plaintext, language, cipher, author, or unique historical production mechanism. The aim is narrower: to define measurement...
Daiki Matsuda· Zenodo (CERN European Organi...· 0 citations
This research technical note describes Union Command Agent, a 6,441,472-parameter encoder-decoder Transformer trained from scratch to translate English and Hinglish device-control requests into a constrained action language. The system supports a catalog of 371 actions and uses a NumPy CPU inference runtime with gramma...
Lalit Belwal· Zenodo (CERN European Organi...· 0 citations
ProCAP is proposed, a probabilistic cross-attentive prompt learning framework that improves cross-modal interaction and training stability without updating any CLIP weights: it learns both visual and textual prompt tokens and links them through stacked bidirectional multi-head cross-attention so the two branches refine...
Hiwa Azeez Abbas, Fatemeh Daneshfar, M. Abdar· 0 citations
This work proposes the Future Representation Action Model (FRAM), a small policy that explicitly links the future end-effector trajectory to the current visual input and selects visual information based on future motion to obtain both high performance and robustness in a small robot policy.
Hiroshi Ito, Hyogo Hiruma, Yoshiki Kanai et al.· 0 citations
A statistical specification for whether valence among the users who write reviews of an app has changed enough to warrant investigation of a release, outage, or support issue is proposed, applied to written Google Play reviews collected under a declared locale, language, sort order, retrieval cap, and window rule.
