Pith. sign in

REVIEW 6 cited by

FreeLB: Enhanced Adversarial Training for Natural Language Understanding

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1909.11764 v5 pith:U36TD6TT submitted 2019-09-25 cs.CL cs.LG

Chen Zhu , Yu Cheng , Zhe Gan , Siqi Sun , Tom Goldstein , Jingjing Liu This is my paper
classification cs.CLcs.LG
keywords adversarialfreelblanguagemodeltrainingapproachbenchmarkexperiments
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Adversarial training, which minimizes the maximal risk for label-preserving input perturbations, has proved to be effective for improving the generalization of language models. In this work, we propose a novel adversarial training algorithm, FreeLB, that promotes higher invariance in the embedding space, by adding adversarial perturbations to word embeddings and minimizing the resultant adversarial risk inside different regions around input samples. To validate the effectiveness of the proposed approach, we apply it to Transformer-based models for natural language understanding and commonsense reasoning tasks. Experiments on the GLUE benchmark show that when applied only to the finetuning stage, it is able to improve the overall test scores of BERT-base model from 78.3 to 79.4, and RoBERTa-large model from 88.5 to 88.8. In addition, the proposed approach achieves state-of-the-art single-model test accuracies of 85.44\% and 67.75\% on ARC-Easy and ARC-Challenge. Experiments on CommonsenseQA benchmark further demonstrate that FreeLB can be generalized and boost the performance of RoBERTa-large model on other tasks as well. Code is available at \url{https://github.com/zhuchen03/FreeLB .

Discussion (0). Sign in to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer

    cs.LG 2019-10 unverdicted novelty 7.0 of 10

    T5 casts all NLP tasks as text-to-text generation, systematically explores pre-training choices, and reaches strong performance on summarization, QA, classification and other tasks via large-scale training on the Colo...

  2. Representation-Guided Parameter-Efficient LLM Unlearning

    cs.CL 2026-04 unverdicted novelty 6.0 of 10

    REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.

  3. Baseline Defenses for Adversarial Attacks Against Aligned Language Models

    cs.LG 2023-09 conditional novelty 6.0 of 10

    Baseline defenses including perplexity-based detection, input preprocessing, and adversarial training offer partial robustness to text adversarial attacks on LLMs, with challenges arising from weak discrete optimizers.

  4. HABIB_TAZ at SemEval-2026 Task 11: Disentangling Formal Logic from Content via Synthetic Training and Multi-Objective Optimization

    cs.CL 2026-07 conditional novelty 5.0 of 10

    A mDeBERTa-v3 system trained on synthetic syllogisms with a debiasing multi-objective loss achieved top ranking scores on three SemEval-2026 subtasks and 6th place on the noisy multilingual subtask.

  5. Cheap Reward Hacking Detection

    cs.LG 2026-06 unverdicted novelty 5.0 of 10

    Small transformer encoder with linear probe detects reward hacking at AUC 0.9467 and TPR@5%FPR 0.8296, matching LLM-as-judge accuracy at ~10000x lower per-trajectory cost.

  6. Understanding and Improving Noisy Embedding Techniques in Instruction Finetuning

    cs.LG 2026-05 unverdicted novelty 5.0 of 10

    SymNoise applies symmetric noise to embeddings during instruction fine-tuning and reports 6.7% higher AlpacaEval scores than NEFTune on LLaMA-2-7B.

Pith tools