Conspiracy theories commonly attribute important events to the actions of powerful and secretive actors. While computational research has largely focused on document-level analyses of conspiracy theories, less attention has been paid to identifying the actors that drive such narratives. We develop annotation guidelines...
Helena Mihaljevi\'c, Jolanda Beer, Mareike Lisker et al.· 0 citations
It is found that fine-grained taxonomic representations substantially improve recall, while simultaneously reducing precision, which highlights both the potential and the remaining limitations of conceptually grounded LLMs for antisemitism detection and reasoning.
Katharina Soemer, Helena Mihaljević· arXiv.org· 0 citations
Speech-LLM integration has shown promising results by leveraging extensive textual pretraining, yet its specific benefits for automatic speech recognition (ASR) remain unclear. We observe that as supervised ASR training data increases, the contribution of LLM priors becomes less evident, and simple speech-text joint tr...
Ruchao Fan, Yiming Wang, Rui Zhao et al.· 0 citations
Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a promising framework for enhancing the reasoning ability of large language models. However, much of the existing work is guided by heuristic intuition, leading to divergent algorithmic choices, even contradictory ones that nevertheless report empiric...
Peidong Wang, Demi Wang, Xufang Luo et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A recent Nature Medicine study reports that general-purpose frontier LLMs outperform specialized retrieval-augmented clinical tools on medical benchmarks, and that retrieval can hurt strong models. We ask the natural follow-up: does structured knowledge-graph (KG) grounding change this, and when does grounding help at...
Large language models (LLMs) have become an effective tool for synthetic data generation, including for low-resource languages, where generated data can improve downstream task performance. Current best-performing approaches typically rely on few-shot prompting with target-language examples, which increases inference c...
Jan Cegin, Daniil Gurgurov, Yusser Al Ghussin et al.· 0 citations
Automatic semantic change detection aims to identify how word meanings shift over time, offering insights into both linguistic and societal change. Despite recent progress in computational lexical semantic change (LSC), existing benchmarks and methods struggle to capture bi-directional semantic change, particularly cas...
While Large Reasoning Models (LRMs) show strong performance in English, they often fail to reason in the language of the query, a phenomenon known as language collapse. Existing RL-based fixes typically add a binary language fidelity reward to the accuracy objective, yet still incur trade-off in accuracy, mid-trace cod...
Embedding benchmarks like MTEB report a single score per model, implicitly treating robustness as a static, scalar property. We argue that embedding robustness is multidimensional, since models respond differently to different types of variation, and requires dynamic evaluation to expose failures hidden by static bench...
While large language models (LLMs) augmented with agentic search capabilities show promise for legal reasoning, they overlook a fundamental constraint that applicable law must match the temporal context of each case, as retroactive application of statutes violates core legal principles and leads to erroneous conclusion...
Wei Fan, Yining Zhou, Mufan Zhang et al.· 0 citations
STOP (Structured On-policy Pruning), an on-policy algorithm for analyzing and pruning long-form reasoning traces, which improves the structural efficiency of generated reasoning, and reallocates reasoning effort away from redundant verification and backtracking toward more productive exploration.
Chenjun Xu, Zhen-Nan Zhou, Zhan Su et al.· arXiv.org· 0 citations
Large vision-language models (LVLMs) suffer from visual ungroundedness: they can produce a fluent, confident, and even correct response driven entirely by language priors, with the image contributing nothing to the prediction. Existing confidence estimation methods cannot detect this, as they observe model behavior und...
Reza Khanmohammadi, Erfan Miahi, Simerjot Kaur et al.· 0 citations
By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.