探索全球前沿学术脉络

SLU-2K: A Question-Based Benchmark for Semantic Evaluation of Sign Language Translation

Sign Language Translation (SLT) is typically evaluated with surface-form metrics such as BLEU and ROUGE, which reward lexical overlap but do not directly measure whether a translation preserves the meaning of the source sign sequence. This is in contrast with the final objective of integrating SLT in assistive technology. In this work, we shift the focus from Sign Language Translation (SLT) to Sign Language Understanding (SLU), with particular emphasis on semantic understanding. Specifically, we evaluate systems based on their ability to correctly recover, from the input video, key semantic aspects of the original sentence, such as actions taking place and facts about people and objects. To enable this evaluation systematically, we propose SLU-2K, a dataset of 2,350 closed-ended video question-answer pairs based on the popular PHOENIX-2014T and CSL-Daily datasets. To obtain SLU-2K, we propose and extensively evaluate an automated data generation pipeline which produces questions across 7 categories, namely actions, locations, numbers, objects, people, time, and weather conditions. We show the potential of SLU-2K by evaluating popular Multimodal Large Language Models (MLLMs) and two representative state-of-the-art systems, MMSTL and SpaMo. Our results show that MLLMs reach near-random performance, highlighting the need for a more systematic integration of SLU in current AI systems. Furthermore, state-of-the-art translation systems carefully fine-tuned on in-domain data still exhibit a substantial semantic gap, with results ranging from 56.7% to 75.2%. These findings suggest that current SLT evaluation protocols overestimate true understanding and that future progress should be measured not only by fluency and n-gram overlap, but also by semantic correctness. Code, prompts, and benchmark files are available at https://github.com/ZenoTsT/SLU-2K

Closing the Calibration Gap in Semantic Caching

Semantic caching cuts LLM inference costs by serving a cached response to semantically similar queries. Standard practice evaluates these systems using PR-AUC, a metric that only measures how well scores rank and ignores whether they are usable at a fixed threshold. We show this mismatch leads to systematically poor deployment choices, as models with the highest PR-AUC are often the worst in operation. We introduce Precision-Cache Hit Ratio (P-CHR) AUC, a cache-aware metric that measures precision across cache utilization levels, and Calibration Retention Rate (CRR), which captures how much offline ranking quality survives at deployment. We decompose the operational gap between offline and deployed quality into a recoverable calibration component and an irreducible structural component fixed by the dataset's positive rate. Our experiments show that the calibration gap is governed by the training objective rather than data scale, and post-hoc calibration only partially closes it. Ultimately, model selection for semantic caching is a calibration problem, not a ranking one, and measuring it is the first step to closing the gap.

LedgerAgent: Structured State for Policy-Adherent Tool-Calling Agents

Policy-adherent tool-calling agents in customer-service domains must maintain task states across turns while calling tools and obeying domain policies. Task states consist of relevant facts, identifiers, constraints, and conditions observed through user interaction and tool calls. In standard agents, task states are not represented separately. Observations, tool returns, and policy instructions are placed in the prompt, leaving agents to reconstruct the relevant states from the prompt each time they decide what to do next. This design makes state management implicit, creating two common failure modes. An agent may retrieve the right facts but later ground its decision in stale, missing, or incorrect information; and a syntactically valid tool call may still violate a domain policy that depends on the current task state. We introduce \textsc{LedgerAgent}, an inference-time method for tool-calling agents that maintains observed task states in a separate ledger and renders the states into the prompt. The ledger is also used to check state-dependent policy constraints before environment-changing tool calls are executed, blocking policy violations. Across four customer-service domains and a mixed panel of open- and closed-weight models, \textsc{LedgerAgent} improves average pass\textasciicircum{}k over a standard prompt-based tool-calling approach, with the largest gains under stricter multi-trial consistency metrics.

Diffusion Language Models: An Experimental Analysis

Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range of tasks. Recently, Diffusion Language Models (DLMs) have emerged as an alternative paradigm that generates text through iterative denoising rather than next-token prediction, allowing parallel refinement of entire sequences. While numerous diffusion-based architectures have been proposed, differences in evaluation protocols, datasets, inference budgets, and generation hyperparameters make it difficult to compare their capabilities and understand the trade-offs they offer. In this work, we present a systematic experimental analysis of modern DLMs. Specifically, we evaluate eight state-of-the-art DLMs across eight benchmarks spanning reasoning, coding, translation, knowledge, and structured problem solving, while explicitly considering both generation quality and computational efficiency. Beyond downstream evaluation, we analyze the impact of key inference-time factors, including denoising steps, context length, block size, and parallel unmasking strategies, and complement large-scale experiments with controlled comparisons of smaller models trained under identical conditions. Our analysis highlights the strengths and limitations of diffusion-based language modeling across different tasks, architectures, and inference budgets. We show that the behavior of DLMs is strongly influenced by generation-time design choices, leading to distinct trade-offs between performance and computational efficiency. Overall, our study provides practical insights into the capabilities and deployment characteristics of contemporary DLMs.

PanRes: A database of latent and acquired antimicrobial resistance allowing 3D-based protein homology search

Antimicrobial resistance databases are central to genomic surveillance, but resistance determinants remain distributed across resources with different scopes, structures, and annotations. We developed PanRes, a curated resistance database of 11,717 genes integrating acquired and latent determinants of antibiotic, biocide, and metal resistance within a unified ontology. We predicted representative protein structures and clustered them by structural similarity, grouping proteins into 598 structurally conserved clusters coherent despite sequence divergence. Their structure-guided alignments were used to build Hidden Markov Models (HMMs) for remote homology search. In wastewater metagenomes from seven European cities, PanRes 3D-based HMMs expanded detection beyond high-confidence BLAST, with 35.2% of retained hits identified only by the HMMs and generally showing greater divergence from known proteins. For beta-lactamases, several proteins retained beta-lactamase-like folds and catalytic geometry despite weak sequence similarity. PanRes is available through an interactive web platform (https://panres.rambio.dk/), a structure-informed resource for exploring the whole resistome.