Skip to content
← Back to explorer

CoGate-LSTM: Prototype-Guided Feature-Space Gating for Mitigating Gradient Dilution in Imbalanced Toxic Comment Classification

Noor Islam S. Mohammad · Oct 19, 2025 · Citations: 0

Automatic Metrics
Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Apr 7, 2026, 1:19 PM

Recent

Extraction refreshed

Apr 13, 2026, 6:36 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.35

Abstract

Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models. We propose CoGate-LSTM, a parameter-efficient recurrent architecture built around a novel cosine-similarity feature gating mechanism that adaptively rescales token embeddings by their directional similarity to a learned toxicity prototype. Unlike token-position attention, the gate emphasizes feature directions most informative for minority toxic classes. The model combines frozen multi-source embeddings (GloVe, FastText, and BERT-CLS), a character-level BiLSTM, embedding-space SMOTE, and weighted focal loss. On the Jigsaw Toxic Comment benchmark, CoGate-LSTM achieves 0.881 macro-F1 (95% CI: [0.873, 0.889]) and 96.0% accuracy, outperforming fine-tuned BERT by 6.9 macro-F1 points (p < 0.001) and XGBoost by 4.7, while using only 7.3M parameters (about 15$\times$ fewer than BERT) and 48 ms CPU inference latency. Gains are strongest on minority labels, with F1 improvements of +71% for severe_toxic, +33% for threat, and +28% for identity_hate relative to fine-tuned BERT. Ablations identify cosine gating as the primary driver of performance (-4.8 macro-F1 when removed), with additional benefits from character-level fusion (-2.4) and multi-head attention (-2.9). CoGate-LSTM also transfers reasonably across datasets, reaching a 0.71 macro-F1 zero-shot on the Contextual Abuse Dataset and 0.73 with lightweight threshold adaptation. These results show that direction-aware feature gating offers an effective and efficient alternative to large, fully fine-tuned transformers for classifying imbalanced toxic comments.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction flags indicate low-signal or possible false-positive protocol mapping.
  • Extraction confidence is 0.35 (below strong-reference threshold).
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

A secondary eval reference to pair with stronger protocol papers.

Main weakness

Extraction flags indicate low-signal or possible false-positive protocol mapping.

Trust level

Low

Eval-Fit Score

0/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Detected

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Persisted extraction missing

No explicit feedback protocol extracted.

Evidence snippet: Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models.

Evaluation Modes

partial

Automatic Metrics

Confidence: Low Source: Persisted extraction evidenced

Includes extracted eval setup.

Evidence snippet: Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models.

Benchmarks / Datasets

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No benchmark anchors detected.

Evidence snippet: Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models.

Reported Metrics

partial

Accuracy, F1, F1 macro, Latency, Toxicity

Confidence: Low Source: Persisted extraction evidenced

Useful for evaluation criteria comparison.

Evidence snippet: We propose CoGate-LSTM, a parameter-efficient recurrent architecture built around a novel cosine-similarity feature gating mechanism that adaptively rescales token embeddings by their directional similarity to a learned toxicity prototype.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Toxic text classification for online moderation remains challenging under extreme class imbalance, where rare but high-risk labels such as threat and severe_toxic are consistently underdetected by conventional models.

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: General
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes: Automatic Metrics
  • Agentic eval: None
  • Quality controls: Not reported
  • Confidence: 0.35
  • Flags: low_signal, possible_false_positive

Protocol And Measurement Signals

Benchmarks / Datasets

No benchmark or dataset names were extracted from the available abstract.

Reported Metrics

accuracyf1f1 macrolatencytoxicity

Research Brief

Deterministic synthesis

We propose CoGate-LSTM, a parameter-efficient recurrent architecture built around a novel cosine-similarity feature gating mechanism that adaptively rescales token embeddings by their directional similarity to a learned toxicity prototype. HFEPX signals include Automatic Metrics with confidence 0.35. Updated from current HFEPX corpus.

Generated Apr 13, 2026, 6:36 AM · Grounded in abstract + metadata only

Key Takeaways

  • We propose CoGate-LSTM, a parameter-efficient recurrent architecture built around a novel cosine-similarity feature gating mechanism that adaptively rescales token embeddings by…
  • On the Jigsaw Toxic Comment benchmark, CoGate-LSTM achieves 0.881 macro-F1 (95% CI: [0.873, 0.889]) and 96.0% accuracy, outperforming fine-tuned BERT by 6.9 macro-F1 points (p <…

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Identify benchmark choices from full text before operationalizing conclusions.
  • Validate metric comparability (accuracy, f1, f1 macro).

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Low-signal flag detected: protocol relevance may be indirect.

Research Summary

Contribution Summary

  • We propose CoGate-LSTM, a parameter-efficient recurrent architecture built around a novel cosine-similarity feature gating mechanism that adaptively rescales token embeddings by their directional similarity to a learned toxicity prototype.
  • On the Jigsaw Toxic Comment benchmark, CoGate-LSTM achieves 0.881 macro-F1 (95% CI: [0.873, 0.889]) and 96.0% accuracy, outperforming fine-tuned BERT by 6.9 macro-F1 points (p < 0.001) and XGBoost by 4.7, while using only 7.3M parameters…
  • Gains are strongest on minority labels, with F1 improvements of +71% for severe_toxic, +33% for threat, and +28% for identity_hate relative to fine-tuned BERT.

Why It Matters For Eval

  • On the Jigsaw Toxic Comment benchmark, CoGate-LSTM achieves 0.881 macro-F1 (95% CI: [0.873, 0.889]) and 96.0% accuracy, outperforming fine-tuned BERT by 6.9 macro-F1 points (p < 0.001) and XGBoost by 4.7, while using only 7.3M parameters…

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Pass: Evaluation mode is explicit

    Detected: Automatic Metrics

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Gap: Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Pass: Metric reporting is present

    Detected: accuracy, f1, f1 macro, latency

Category-Adjacent Papers (Broader Context)

These papers are nearby in arXiv category and useful for broader context, but not necessarily protocol-matched to this paper.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote