Machine Translation Digest for Jun 01 2025
Here is today's selection of cs.CL papers focusing on advancements in language model evaluation and cultural nuances in language processing. The papers explore innovative approaches to enhance the performance and evaluation of large language models, particularly emphasizing the importance of cultural context and adversarial testing in improving model accuracy and reliability.
COMPKE: Complex Question Answering under Knowledge Editing
Knowledge Editing, which efficiently modifies the knowledge in large language models, has gathered great attention. Current benchmarks primarily use multi-hop question answering to assess and analyze newly injected or updated knowledge. However, we argue that these benchmarks fail to effectively evaluate how well the updated models apply this knowledge in real-life scenarios, particularly when questions require complex reasoning, involving one-to-many relationships or multi-step logical intersections. To fill in this gap, we introduce a new benchmark, COMPKE: Complex Question Answering under Knowledge Editing, which includes 11,924 complex questions that reflect real-life situations. We conduct an extensive evaluation of four knowledge editing methods on COMPKE, revealing that their effectiveness varies notably across different models. For instance, MeLLo attains an accuracy of 39.47 on GPT-4O-MINI, but this drops sharply to 3.83 on QWEN2.5-3B. We further investigate the underlying causes of these disparities from both methodological and model-specific perspectives. The datasets are available at https://github.com/kzjkzj666/CompKE.
Improving Automatic Evaluation of Large Language Models (LLMs) in Biomedical Relation Extraction via LLMs-as-the-Judge
Large Language Models (LLMs) have demonstrated impressive performance in biomedical relation extraction, even in zero-shot scenarios. However, evaluating LLMs in this task remains challenging due to their ability to generate human-like text, often producing synonyms or abbreviations of gold-standard answers, making traditional automatic evaluation metrics unreliable. On the other hand, while human evaluation is more reliable, it is costly and time-consuming, making it impractical for real-world applications. This paper investigates the use of LLMs-as-the-Judge as an alternative evaluation method for biomedical relation extraction. We benchmark 8 LLMs as judges to evaluate the responses generated by 5 other LLMs across 3 biomedical relation extraction datasets. Unlike other text-generation tasks, we observe that LLM-based judges perform quite poorly (usually below 50% accuracy) in the biomedical relation extraction task. Our findings reveal that it happens mainly because relations extracted by LLMs do not adhere to any standard format. To address this, we propose structured output formatting for LLM-generated responses that helps LLM-Judges to improve their performance by about 15% (on average). We also introduce a domain adaptation technique to further enhance LLM-Judge performance by effectively transferring knowledge between datasets. We release both our human-annotated and LLM-annotated judgment data (36k samples in total) for public use here: https://github.com/tahmedge/llm_judge_biomedical_re.
Trick or Neat: Adversarial Ambiguity and Language Model Evaluation
Detecting ambiguity is important for language understanding, including uncertainty estimation, humour detection, and processing garden path sentences. We assess language models' sensitivity to ambiguity by introducing an adversarial ambiguity dataset that includes syntactic, lexical, and phonological ambiguities along with adversarial variations (e.g., word-order changes, synonym replacements, and random-based alterations). Our findings show that direct prompting fails to robustly identify ambiguity, while linear probes trained on model representations can decode ambiguity with high accuracy, sometimes exceeding 90\%. Our results offer insights into the prompting paradigm and how language models encode ambiguity at different layers. We release both our code and data: https://github.com/coastalcph/lm_ambiguity.
Culturally-Grounded Chain-of-Thought (CG-CoT):Enhancing LLM Performance on Culturally-Specific Tasks in Low-Resource Languages
Large Language Models (LLMs) struggle with culturally-specific reasoning tasks, particularly in low-resource languages, hindering their global applicability. Addressing this gap is crucial for equitable AI deployment. We introduce Culturally-Grounded Chain-of-Thought (CG-CoT), a novel prompting strategy that combines dense vector retrieval of cultural context with explicit reasoning sequences. Our extensive experiments on Yoruba proverb interpretation demonstrate that CG-CoT provides significantly higher culturally-aligned accuracy and depth than traditional prompting methods, validated through both automated metrics and LLM-based evaluations. Notably, we uncover stark disparities between token-level translation metrics like BLEU and human-judged cultural relevance, suggesting a rethinking of evaluation approaches for low-resource NLP.
LAQuer: Localized Attribution Queries in Content-grounded Generation
Grounded text generation models often produce content that deviates from their source material, requiring user verification to ensure accuracy. Existing attribution methods associate entire sentences with source documents, which can be overwhelming for users seeking to fact-check specific claims. In contrast, existing sub-sentence attribution methods may be more precise but fail to align with users' interests. In light of these limitations, we introduce Localized Attribution Queries (LAQuer), a new task that localizes selected spans of generated output to their corresponding source spans, allowing fine-grained and user-directed attribution. We compare two approaches for the LAQuer task, including prompting large language models (LLMs) and leveraging LLM internal representations. We then explore a modeling framework that extends existing attributed text generation methods to LAQuer. We evaluate this framework across two grounded text generation tasks: Multi-document Summarization (MDS) and Long-form Question Answering (LFQA). Our findings show that LAQuer methods significantly reduce the length of the attributed text. Our contributions include: (1) proposing the LAQuer task to enhance attribution usability, (2) suggesting a modeling framework and benchmarking multiple baselines, and (3) proposing a new evaluation setting to promote future research on localized attribution in content-grounded generation.
| Curated by yukajii.com |