BatCheet is introduced, a multilingual dialogue corpus named after the Hindi term for conversation or chitchat, containing approximately 49,000 dialogues for dialogue translation across five translation directions and fine-tuning five open-source LLMs yields substantial gains over zero- and few-shot baselines.
Priyanka Dasari, Yuvrajsinh Bodana, Vandan Mujadia et al.· Proceedings of the Eleventh...· 0 citations
Reinforcement learning with binary correctness rewards trains correctness, not calibrated confidence. The confidence that reasoning models verbalize is systematically overconfident, and the problem is not merely one of scale: verbalized confidence tracks how willing a model is to commit to an answer, not how likely the...
Ya-Dong Xi, Rong-Sheng Zhang, Tang-Jie Lv et al.· 0 citations
With this ensemble, frequency matching with a public name list recovers zero identities by alignment across the four corpora; the names it got right were ones the detector missed, left in clear text.
Andreas K. Maier, Monica Hinrichs-Mayer, Franziska Weber et al.· 0 citations
This work proposes Threader, a memory system that shifts the focus from memory construction to efficient, structure-aware access over raw interactions, and preserves them as first-class memory, organizes them into topic-coherent segments via lightweight incremental segmentation, and enables accurate retrieval through m...
Dong-Hua Cai, Yong-Heng Deng, Yi-Fei Wang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
CoLMbo-SV, a speaker language model that combines strong speaker discrimination with structured, acoustically grounded comparison reports, substantially advance audio-language speaker verification, bring its accuracy toward that of dedicated speaker encoders while adding checkable acoustic reporting, and establish an e...
Massa Baali, Sarthak Bisht, Zi-Yue Qiu et al.· 0 citations
MULTISPEECHQA, a large-scale, synthetically generated and human-verified dataset comprising 9200 hours of 10.8 million spoken question-answer pairs in 23 typologically diverse languages, is presented and shows that high-quality synthetic datasets offer a cheap solution to improving the multilingual capabilities of SLMs...
Tolúlopé Ògúnrèmí, D. Jurafsky, Christopher D. Manning et al.· Interspeech· 0 citations
Test-time scaling and post-training have improved LLM performance in coding and mathematical reasoning, but their effectiveness for individual stance prediction remains unclear. We study this question by predicting a person's stance in a new discussion from their history. We evaluate widely used test-time scaling strat...
Yuyang Zhao, Xuan Liu, HaoYang Shangm Haojian Jin· 0 citations
This suite provides a new efficient lens on generalization: select intermediate pre-training checkpoints that strongly generalize reasoning and alignment, better than the final pre- or mid-training checkpoints, and select pre-training data that controls and stabilizes generalization dynamics.
Jia-Xin Wen, Zhengxuan Wu, D. Song et al.· 0 citations
This work separate factual recall into three steps: generating a correct candidate, ranking the available candidates, and selecting the final answer, and shows that these capabilities do not collapse into a single notion of ``knowing''.
This work proposes an alternative approach for identifying a sentence's past, present, or future global reference interval by encoding sentences using fixed-length inventory vectors that are comprised of part-of-speech counts, dependency relation counts, and explicit futurate pattern counts.
This work shows that LLM-as-Judge is operationalized on cloud serving infrastructure, not of any particular model family, and proposes a minimal, low-cost reporting protocol: several judge re-runs, published stability profiles and adjacency intervals, and results under at least two judges from different families.
This work measures knowledge capacity, the factual bits stored per parameter, on synthetic biographies with known information content and suggests judging low-precision training by retained capacity during training rather than final loss.
Prashanna Mani Paudel, S. Sheshappanavar· 0 citations
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.
