arXiv:2604.01306v1 Announce Type: new
Abstract: Evaluating scientific arguments requires assessing the strict consistency between a claim and its underlying multimodal evidence. However, existing benchmarks lack the scale, domain diversity, and visual…
arXiv:2603.22862v2 Announce Type: replace-cross
Abstract: Tool use enables large language models (LLMs) to access external information, invoke software systems, and act in digital environments beyond what can be solved from model parameters alone. Ear…
arXiv:2603.20673v2 Announce Type: replace
Abstract: Retrieval-augmented language models can retrieve relevant evidence yet still commit to answers before explicitly checking whether the retrieved context supports the conclusion. We present PAVE (Premi…
arXiv:2604.02118v1 Announce Type: cross
Abstract: Evaluating factual correctness of LLM generated natural language explanations grounded in time series data remains an open challenge. Although modern models generate textual interpretations of numerica…
arXiv:2601.17387v2 Announce Type: replace
Abstract: Multilingual speech-text models rely on cross-modal language alignment to transfer knowledge between speech and text, but it remains unclear whether this reflects shared computation for the same lang…
arXiv:2604.01624v1 Announce Type: cross
Abstract: Diffusion language models (DLMs) expose their denoising trajectories, offering a natural handle for inference-time control; accordingly, an ideal hallucination mitigation framework should intervene dur…
arXiv:2510.27176v4 Announce Type: replace-cross
Abstract: Can AI autonomously design mechanisms for computer systems on par with the creativity and reasoning of human experts? We present Glia, an AI architecture for networked systems design that uses …
arXiv:2604.01538v1 Announce Type: new
Abstract: Large language models have been adopted in the medical domain for clinical documentation to reduce clinician burden. However, studies have reported that LLMs often “forget” a significant amount of instru…
arXiv:2604.01707v1 Announce Type: new
Abstract: Memory emerges as the core module in the large language model (LLM)-based agents for long-horizon complex tasks (e.g., multi-turn dialogue, game playing, scientific discovery), where memory can enable kn…
arXiv:2604.01957v1 Announce Type: new
Abstract: Machine-translated benchmark datasets reduce costs and offer scale, but noise, loss of structure, and uneven quality weaken confidence. What matters is not merely whether we can translate, but also wheth…