Conversational database interfaces face a critical challenge: users naturally embed queries in conversational noise (greetings, politeness, off-topic remarks), which degrades intent classification accuracy and wastes computational resources. Despite advances in orchestration and retrieval strategies, a fundamental ques...
Harness-Search is introduced, a multi-agent search harness to reduce the local errors propagating across subsequent exploration, evidence curation, and termination decisions, and improves both retrieval and answer generation under the same policy backbone.
Shanyong Wang, Zhen-Wen Ji, Lei Jin et al.· 0 citations
On-policy distillation (OPD) has emerged as an effective post-training paradigm for language models, with recent efforts extending it to block diffusion language models (BDLMs). However, existing studies focus almost exclusively on small block sizes, leaving distillation into student models with larger blocks underexpl...
Zai-Quan Yang, Fei Wei, Yong Wang et al.· 0 citations
\`{I}r\`{o}y\`{i}nSpeech is a 42-hour, 80-speaker Yor\`ub\'a read-speech corpus whose audio has been distributed by ELRA since 2024. This paper describes the release of its text component: 24,905 unique, hand-verified, tone-marked Yor\`ub\'a sentences (275,897 tokens; 15,687 types), curated in 2022 as recording prompts...
Kola Tubosun, Aanuoluwapo Aremu, Tolulope Ogunremi et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
An adequate conversational answer may differ from a short or incomplete benchmark reference. To measure adequacy against the recorded history we prefer source-aware grading, in which the judge checks the reference against the full source before assessing system-blinded answers; original reference-only grading is report...
Rubric-based reinforcement learning (RL) provides interpretable rewards for aligning large language models (LLMs) by evaluating responses against query-specific evaluation criteria. To construct rubrics at scale, a straightforward approach to LLM-based rubric generation is to prompt an LLM to generate a rubric directly...
Haocheng Yang, Yuchao Zhang, Licheng Pan et al.· 0 citations
Red-TTT is proposed, which updates the attacker's parameters during the attack on each behavior, and requires only sampling access to the victim and integrates into existing attack pipelines with no other changes, increasing attack success rate and cracking many behaviors previous method cannot.
Tong-Yan Hu, Hao Li, Xiaogeng Liu et al.· 0 citations
A large part of political discourse on social media is built and expressed at a level of claims: i.e. declarative, typically single-clause statements, which convey a particular interpretation of reality and can range from factual to evaluative. Moreover, rather than occurring randomly, claims coalesce, recur in pattern...
Test-time compute has become a central way to improve code generation: systems sample multiple candidate programs and use verifier-visible evidence to select the final output. This paradigm implicitly assumes that the verifier provides a corrective signal independent from the generator. We challenge this assumption und...
Feng He, He-Jia Wang, Ling-Hao Meng et al.· 0 citations
The repetition count that works best for a small language model may not remain best at a larger scale. We study this effect in pretraining with a finite target corpus mixed with generic data at a fixed target fraction. On Wikipedia-derived data and Proof-Pile-2, the ranking of measured repetition counts changes with mo...
Large language models (LLMs) progressively revise their predictions across Transformer layers, yet we typically observe only the final output, discarding the trajectory through which it is formed. We introduce Belief-Trajectory Energy(BTE), a model-grounded measure that characterizes an input through the layerwise pred...
Multimodal inputs are inherently heterogeneous, not only across modalities but also in the information pathways required for effective prediction. To address this limitation, we propose InstMoE, an adaptive expert routing framework for multimodal learning. InstMoE dynamically routes each input to specialized unimodal a...
Gui-Min Hu, Xiang He, Ying-Jian Li 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.
