Test-time training (TTT) lets a model store information in its weights during inference. When the model learns from its own output, however, each update also changes the model that generates the next training example. Across 128K-token streams, retaining generated-text updates worsens prediction on independent human-wr...
Cheng-Zhi Luo, Bing Li, Bernard Ghanem· 0 citations
State-of-the-art ASR systems primarily target native adult speech, leading to substantial performance gaps for children, adolescents, and L2 speakers. We introduce AraYoungVoices, a 151.72-hour Arabic read-speech corpus from 286 speakers aged 7--18, comprising AraKids (7--12) and AraTeens (13--18). The corpus includes...
S. Chowdhury, Zien Sheikh Ali, H. Lachemat et al.· 1 citation
When does communication help a group identify hidden adversaries, and how does its value change as the group adapts? In Mafia, an informed minority hides inside an uninformed majority whose only evidence is open play. The zero-information game, where each day's vote eliminates a random player, is exactly solved and sco...
Work agents operate over digital artifacts to execute professional knowledge-intensive work, requiring training environments that support long-horizon interaction and trustworthy verification. However, hand-crafted environments incur prohibitive engineering overhead that prevents environment scaling, whereas synthesis...
Jiazheng Zhang, Long Ma, Yunxian Yang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
The results show that surprisal is predictive of Chinese reading times, and it is found that checkpoints whose surprisal remains closer to $n$-gram statistics are better predictors of reading in HKP and MECO, providing one possible explanation for the inverse scaling behavior.
Hon-Gao Zhu, Mu-Di Xu, Yikang Liu et al.· 0 citations
This work introduces SilK (Silhouette-guided K-means), which trains on a Cluster Validation Index, an internal measure of cluster quality without using labels, and shows how this method stays competitive with the best baselines on three downstream tasks.
Kishor Kumar Bhaumik, Nícolas Roque dos Santos, Neil Shah et al.· 0 citations
Findings show that without external verification, supervision is itself a critical trust boundary: systems must distinguish abstention from rejection, preserve useful disagreement, and preserve the provenance and polarity of information before it becomes future context.
Autoregressive generation in Large Language Models (LLMs) is constrained by the memory and computational demands of attention mechanisms. Sparse attention methods mitigate this cost by selecting only high-probability entries of the attention matrix. We observe that in many such methods, this renders the probability-val...
Noam Elata, Itay Lamprecht, Mikey Shechter et al.· 0 citations
Modulation is introduced as a setting where the persona context is already embedded in the content being manipulated, requiring control methods to amplify or suppress a trait already present rather than inject it from scratch.
Activation probes can predict safety-relevant properties of language models with high area under the receiver-operating-characteristic curve (AUROC), but deployed agent monitors make thresholded alarm decisions under tight false-alarm budgets. These are different estimands. We introduce an Operational Validity Contract...
Large language models (LLMs) increasingly serve as autonomous agents that invoke external tools. However, this capability introduces tool hallucination, selecting incorrect tools or generating invalid calls. Existing mitigation methods report substantial improvements, yet we identify a previously overlooked failure mod...
Peigui Qi, Kunsheng Tang, Yide Song et al.· 0 citations
General turn-taking behavior in real-time dialogue systems requires deciding whether to keep listening or start responding while listening, and whether to continue or stop while speaking. Existing turn detectors use heterogeneous, task-specific label spaces and are often trained on limited annotations or evaluated on i...
Zhan-Xun Liu, Yi-Fan Duan, Heng-Tao Wu et al.· 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.
