A Reproducible Multi-Architecture Baseline for Token-Level Chinese Metaphor Identification under the MIPVU Framework

• URL: http://arxiv.org/abs/2605.07170v1
• Date: Fri, 08 May 2026 03:07:19 GMT
• Title: A Reproducible Multi-Architecture Baseline for Token-Level Chinese Metaphor Identification under the MIPVU Framework
• Abstract summary: This paper presents a reproducible multi-architecture baseline for token-level metaphor identification on the PSU Chinese Metaphor Corpus (PSU CMC)<n>We compare three model families: encoder fine-tuning with Chinese RoBERTa-wwm-ext-large; (ii) MelBERT adapted to Chinese using a newly constructed basic-meaning resource derived from the Modern Chinese Dictionary, 7th edition (MCD7)<n>Across five fixed seeds, MelBERT MIP-only achieves the strongest performance at 0.7281 +/- 0.0050 test positive F1, marginally above MelBERT Full (0.7270 +/- 0.0069) and clearly above plain RoBERTa (0.7
• Score: 2.545461559283292
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Metaphor is pervasive in everyday language, yet token-level computational identification of metaphor-related words in Chinese under the MIPVU framework remains under-explored relative to English. This paper presents a reproducible multi-architecture baseline for token-level metaphor identification on the PSU Chinese Metaphor Corpus (PSU CMC), the only widely available MIPVU-annotated Chinese corpus. We systematically compare three model families: (i) encoder fine-tuning with Chinese RoBERTa-wwm-ext-large; (ii) MelBERT adapted to Chinese using a newly constructed basic-meaning resource derived from the Modern Chinese Dictionary, 7th edition (MCD7), comprising 74,823 entries with 71.51% PSU CMC vocabulary coverage; and (iii) Qwen3.5-9B fine-tuned with QLoRA as an instruction-tuned generative baseline. Across five fixed seeds, MelBERT MIP-only achieves the strongest performance at 0.7281 +/- 0.0050 test positive F1, marginally above MelBERT Full (0.7270 +/- 0.0069) and clearly above plain RoBERTa (0.7142 +/- 0.0121). The Qwen QLoRA generative configuration trails encoder baselines by approximately 11 F1 points (0.6157 +/- 0.0113). Three findings merit attention: (1) the SPV channel of MelBERT does not contribute reliable positive signal in Chinese, consistent with the dominance of conventional metaphor; (2) the Qwen-encoder gap is concentrated in recall, reflecting the discrete-commitment limitation of generative output; (3) several Qwen task formulations fail due to format design rather than model capacity. We release all split manifests, per-seed outputs, the MCD7 basic-meaning embedding pipeline, and training scripts to serve as a common reference for future Chinese metaphor identification research.

Related papers

• Interpretable Chinese Metaphor Identification via LLM-Assisted MIPVU Rule Script Generation: A Comparative Protocol Study [1.482489178857542]
  We present an LLM-assisted pipeline that operationalises four metaphor identification protocols.<n>Each protocol is a modular chain of deterministic steps interleaved with controlled LLM calls.<n>We evaluate on seven Chinese metaphor datasets spanning token-, sentence-, and span-level annotation.
  arXiv  Detail & Related papers  (2026-03-11T13:55:34Z)
• PolyFrame at MWE-2026 AdMIRe 2: When Words Are Not Enough: Multimodal Idiom Disambiguation [0.533024001730262]
  PolyFrame is a unified pipeline for both image+text ranking (Subtask A) and text-only caption ranking (Subtask B)<n>All model variants retain frozen CLIP-style vision-language encoders and the multilingual BGE M3 encoder, training only lightweight modules.<n>On the multilingual blind test, our systems achieved average Top-1/NDCG scores of 0.35/0.73 for Subtask A and 0.32/0.71 for Subtask B across 15 languages.
  arXiv  Detail & Related papers  (2026-02-20T23:07:55Z)
• Qwen3-ASR Technical Report [71.87071808763484]
  We introduce Qwen3-ASR family, which includes two powerful all-in-one speech recognition models and a novel non-autoregressive speech forced alignment model.<n>Qwen3-ASR-1.7B and Qwen3-ASR-0.6B are ASR models that support language identification and ASR for 52 languages and dialects.
  arXiv  Detail & Related papers  (2026-01-29T06:58:13Z)
• Comparative Study of Large Language Models on Chinese Film Script Continuation: An Empirical Analysis Based on GPT-5.2 and Qwen-Max [6.2773052078641385]
  This study constructs the first Chinese film script continuation benchmark comprising 53 classic films.<n> Evaluation integrates ROUGE-L, Structural Similarity, and LLM-as-Judge scoring.<n>GPT-5.2 excels in character consistency, tone-style matching, and format preservation, while Qwen-Max shows deficiencies in generation stability.
  arXiv  Detail & Related papers  (2026-01-21T09:55:26Z)
• Chinese ModernBERT with Whole-Word Masking [15.633276132474316]
  We introduce Chinese ModernBERT, a hardware-aware 32k BPE vocabulary tailored to frequent Chinese affixes/compounds.<n>We pre-train on 1.2T Chinese tokens from CCI3-HQ, CCI4 (Chinese), and Cosmopedia-Chinese.<n>On CLUE, Chinese ModernBERT is competitive with strong Chinese encoders under a unified fine-tuning protocol.
  arXiv  Detail & Related papers  (2025-10-14T08:41:22Z)
• MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipes [60.57770396565211]
  We show that strong reasoning abilities can emerge with far less data.<n>MobileLLM-R50M achieves an AIME score of 15.5, compared to just 0.6 for OLMo-2-1.48B and 0.3 for SmolLM-2-1.7B.
  arXiv  Detail & Related papers  (2025-09-29T15:43:59Z)
• XplaiNLP at CheckThat! 2025: Multilingual Subjectivity Detection with Finetuned Transformers and Prompt-Based Inference with Large Language Models [2.749729059235755]
  This notebook reports the Xplai submission to the CheckThat! 2025 shared task on multilingual subjectivity detection.<n>We evaluate two approaches: supervised fine-tuning of transformer encoders, EuroBERT, XLM-RoBERTa, and German-BERT, on monolingual and machine-translated training data.<n>For German, a German-BERT model fine-tuned on translated training data from typologically related languages yields competitive performance over the baseline.
  arXiv  Detail & Related papers  (2025-09-15T16:53:41Z)
• HeQ: a Large and Diverse Hebrew Reading Comprehension Benchmark [54.73504952691398]
  We set out to deliver a Hebrew Machine Reading dataset as extractive Questioning.<n>The morphologically rich nature of Hebrew poses a challenge to this endeavor.<n>We devise a novel set of guidelines, a controlled crowdsourcing protocol, and revised evaluation metrics.
  arXiv  Detail & Related papers  (2025-08-03T15:53:01Z)
• MonkeyOCR: Document Parsing with a Structure-Recognition-Relation Triplet Paradigm [60.14048367611333]
  MonkeyOCR is a vision-language model for document parsing.<n>It advances the state of the art by leveraging a Structure-Recognition-Relation (SRR) triplet paradigm.
  arXiv  Detail & Related papers  (2025-06-05T16:34:57Z)
• Youku-mPLUG: A 10 Million Large-scale Chinese Video-Language Dataset for Pre-training and Benchmarks [63.09588102724274]
  We release the largest public Chinese high-quality video-language dataset named Youku-mPLUG. Youku-mPLUG contains 10 million Chinese video-text pairs filtered from 400 million raw videos across a wide range of 45 diverse categories for large-scale pre-training. We build the largest human-annotated Chinese benchmarks covering three popular video-language tasks of cross-modal retrieval, video captioning, and video category classification.
  arXiv  Detail & Related papers  (2023-06-07T11:52:36Z)
• CPM: A Large-scale Generative Chinese Pre-trained Language Model [76.65305358932393]
  We release the Chinese Pre-trained Language Model (CPM) with generative pre-training on large-scale Chinese training data. CPM achieves strong performance on many NLP tasks in the settings of few-shot (even zero-shot) learning.
  arXiv  Detail & Related papers  (2020-12-01T11:32:56Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.