Structuring Semantic Embeddings for Principle Evaluation: A Prototype-Guided Contrastive Learning Approach

• URL: http://arxiv.org/abs/2608.15224v1
• Date: Sat, 15 Aug 2026 13:19:25 GMT
• Title: Structuring Semantic Embeddings for Principle Evaluation: A Prototype-Guided Contrastive Learning Approach
• Abstract summary: General-purpose text embeddings are widely deployed for such tasks, but broad semantic similarity can place semantically similar yet task-distinct examples.<n>We introduce Prototype-Guided Contrastive Learning (PGCL), a prototype-guided geometric regularization module built on top of frozen text embeddings.
• Score: 58.15427952222424
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reliable post-hoc evaluation asks whether already generated text satisfies a target criterion after generation. In this paper we study a focused frozen-embedding setting using principle-evaluation proxy tasks: toxicity detection, fine-grained emotion categorization, and ordinal review rating. General-purpose text embeddings are widely deployed for such tasks, but broad semantic similarity can place semantically similar yet task-distinct examples in overlapping regions of the representation space. We introduce Prototype-Guided Contrastive Learning (PGCL), a prototype-guided geometric regularization module built on top of frozen text embeddings. The module combines a semantic stream, a prototype-anchor attention stream, supervised contrastive learning, offset-based prototype-margin regularization, and stream regularization to produce a compact task-adapted representation without updating the base encoder. Controlled experiments show that PGCL improves over raw frozen embeddings on all three datasets and gives the clearest direct-baseline margin on AmazonReviews, while remaining competitive with strong direct frozen metric-learning baselines on GoEmotions and ToxicComment. We also add supervised residual-adapter, encoder-LoRA, full fine-tuning, objective ablation, sensitivity, and fully logged few-shot LLM protocol diagnostics to define the boundary of the claim. The theoretical analysis is revised as a sufficient-condition account for prototype-margin behavior under explicit assumptions in the prototype-mapping space, rather than as an unconditional training or final-embedding separation guarantee.

Related papers

• Toward Robust LiDAR Semantic Segmentation for Real-World Deployment: Evaluation under Coarse Labels, Adverse Conditions, and Domain Shifts [39.58684038370709]
  LiDAR-based semantic segmentation is a core perception module for autonomous vehicles and mobile robots.<n>Existing evaluation protocols focus on clean, single-domain settings and fine-grained label robustness, leaving deployment readiness largely unassessed.<n>We propose a structured evaluation protocol that assesses the deployment readiness of LiDAR semantic segmentation models along three complementary dimensions.
  arXiv  Detail & Related papers  (2026-09-02T17:15:07Z)
• What to Test Next: Interpretable Coverage Gap Discovery in Driving VLMs [52.50210189669399]
  Driving vision-language models (VLMs) must accurately understand scenes across diverse conditions defined by Operational Design Domains (ODDs)<n>We propose SliceScorer, a deterministic scoring rule for missing-slice recommendation.<n>We show that SliceNav surfaces high-risk coverage gaps more effectively than prior slice-discovery methods.
  arXiv  Detail & Related papers  (2026-06-01T03:18:01Z)
• Hierarchical Consistency Learning for Test-time Adaptation in Camouflage Perception [50.278200968044665]
  Camouflaged object detection (COD) aims to localize targets that exhibit minimal perceptual differences from backgrounds through physical attributes.<n>Existing methods, constrained by the static train-then-freeze paradigm, suffer from domain rigidity and dependency annotation.<n>We propose the hierarchical consistency learning framework, which integrates test-time adaptation for dynamic representation recalibration.
  arXiv  Detail & Related papers  (2026-05-25T09:57:46Z)
• RAGR: Review-Augmented Generative Recommendation [61.29879991713178]
  Review-Augmented Generative Recommendation (RAGR) is a novel GR framework that incorporates review feedback directly into the generative user sequence.<n>RAGR yields consistent and significant gains over strong GR backbones across all metrics.
  arXiv  Detail & Related papers  (2026-05-17T05:21:23Z)
• The Compliance Paradox: Semantic-Instruction Decoupling in Automated Academic Code Evaluation [11.984098021215878]
  We introduce the Semantic-Preserving Adrial Code Injection (SPACI) Framework and the Abstract Syntax Tree-Aware Semantic Injection Protocol (AST-ASIP)<n>These methods exploit the Syntax-Semantics Gap by embedding adversarial directives into syntactically inert regions (trivia nodes) of the Abstract Syntax Tree.<n>Through a large-scale evaluation of 9 SOTA models across 25,000 submissions in Python, C, C++, and Java, we reveal catastrophic failure rates (>95%) in high-capacity open-weights models like DeepSeek-V3.
  arXiv  Detail & Related papers  (2026-01-29T07:40:58Z)
• Unsupervised Text Segmentation via Kernel Change-Point Detection on Sentence Embeddings [0.0]
  Unsupervised text segmentation is crucial because boundary labels are expensive, subjective, and often fail to transfer across domains and granularity choices.<n>We propose Embed-KCPD, a training-free method that represents sentences as embedding vectors and estimates boundaries by minimizing a penalized KCPD objective.<n>A case study on Taylor Swift's tweets illustrates that Embed-KCPD combines strong theoretical guarantees, simulated reliability, and practical effectiveness for text segmentation.
  arXiv  Detail & Related papers  (2026-01-26T18:54:34Z)
• Prototypical Contrastive Learning through Alignment and Uniformity for Recommendation [6.790779112538357]
  We present underlinePrototypical contrastive learning through underlineAlignment and underlineUniformity for recommendation. Specifically, we first propose prototypes as a latent space to ensure consistency across different augmentations from the origin graph. The absence of explicit negatives means that directly optimizing the consistency loss between instance and prototype could easily result in dimensional collapse issues.
  arXiv  Detail & Related papers  (2024-02-03T08:19:26Z)
• AMRFact: Enhancing Summarization Factuality Evaluation with AMR-Driven Negative Samples Generation [57.8363998797433]
  We propose AMRFact, a framework that generates perturbed summaries using Abstract Meaning Representations (AMRs) Our approach parses factually consistent summaries into AMR graphs and injects controlled factual inconsistencies to create negative examples, allowing for coherent factually inconsistent summaries to be generated with high error-type coverage.
  arXiv  Detail & Related papers  (2023-11-16T02:56:29Z)
• Uncovering Prototypical Knowledge for Weakly Open-Vocabulary Semantic Segmentation [59.37587762543934]
  This paper studies the problem of weakly open-vocabulary semantic segmentation (WOVSS) Existing methods suffer from a granularity inconsistency regarding the usage of group tokens. We propose the prototypical guidance network (PGSeg) that incorporates multi-modal regularization.
  arXiv  Detail & Related papers  (2023-10-29T13:18:00Z)
• Prototypical Kernel Learning and Open-set Foreground Perception for Generalized Few-shot Semantic Segmentation [7.707161030443157]
  Generalized Few-shot Semantic (GFSS) extends Few-shot Semantic aggregation to segment unseen classes and seen classes during evaluation. We address the aforementioned problems by jointing the prototypical kernel learning and open-set perception. In addition, a foreground contextual perception module cooperating with conditional bias based inference is adopted to perform class-agnostic as well as open-set foreground detection.
  arXiv  Detail & Related papers  (2023-08-09T13:38:52Z)

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.