Experiments with silenced and replacement audio suggest that RT-OPD strengthens the student's reliance on acoustic evidence, and the 3B model achieves 72.72% accuracy on MMAU, the highest among the compared 3B models and competitive with several 7B and 8B models.
A local-official discrepancy in metric scale and checkpoint ranking is analyzed, highlighting limitations of in-distribution diagnostics for predicting Dynabench probe outcomes and a local-official discrepancy in metric scale and checkpoint ranking is analyzed.
P. Subedi, Howard Prioleau, Saurav K. Aryal· 0 citations
Results show that BLV-focused post-training improves both accessibility-specific spatial grounding and general visual-text reasoning, and is substantially outperforms the baseline across various benchmarks.
Rishabh C. Choudhary, S. Raj, Umesh Goyal et al.· 0 citations
This study examines the Voxtral audio-language model (ALM) framework for spoofing detection, as a step toward combining CM capabilities within the ALM framework, and proposes an instruction-guided approach that uses label-sequence likelihoods to evaluate bonafide and spoofed speech.
Avishai Weizman, Yehuda Ben-Shimol, I. Lapidot· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A three-stage audit of six widely accessible conversational AI systems compares refusal behaviour across 1,600 crossed prompts per system, isolates relational framing through 300 matched prompt pairs, and contrasts pre-submission framing with post-output critique across fresh sessions.
This work uncovers short context additions that redirect 61.4%-73.2% of each decision system's initially correct decisions toward a wrong option fixed in advance, exposing a weakness in current decision models.
Zi-Xiang Xu, Zirui Song, Chi-Yu Zhang et al.· 0 citations
SemMSA, a latent semantic-aided framework that constructs rich sentiment-relevant semantics with LLMs, fully integrating with all modalities via anchor-free spectral alignment, is proposed.
Wen-Hao Li, Zhi-Bin Wu, Chong-Yao Xiao et al.· 0 citations
ARGUS, a source-grounded pipeline that combines a 5W1H-inspired schema, legal-domain models, and LLM-based structured generation to construct document-level Event Knowledge Graphs (EKGs) from CourtListener complaints, suggests that EKGs are most useful for organizing and reasoning over evidence once relevant material h...
Automatic speech recognition is typically trained assuming that the reference transcript is the only valid labeling of an utterance, yet even nominally verbatim transcripts contain localized differences in pronunciation, spelling, or lexical realization that the acoustics do not uniquely determine. Omni-temporal Classi...
Saurabh Kumar, Diptiman Mohanta, P. Ghosh· arXiv.org· 0 citations
This hybrid system uses large language models to propose and ground visual elements, estimate their salience to the narrative, and generate compressed realizations, which makes better decisions than prompted LLMs about what to describe and when to describe it.
Igor Sterner, Mirella Lapata, A. Lascarides et al.· 0 citations
A native-reference phone-class geometry for measuring second language (L2) pronunciation deviation without requiring pronunciation labels, read-aloud prompts, or matched recordings from native and L2 speakers is proposed.
Tina Raissi, Nhan Phan, Chen-Xi Wang et al.· 0 citations
The failure modes for the parsing tasks are characterized, and it is concluded that a generation score alone does not show that models understand formal semantic representations.
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
