pith. sign in

arxiv: 1905.05950 · v2 · pith:GLGB4JDZnew · submitted 2019-05-15 · 💻 cs.CL

BERT Rediscovers the Classical NLP Pipeline

classification 💻 cs.CL
keywords modelpipelinebertinformationadjustadvancedanalysisappear
0
0 comments X
read the original abstract

Pre-trained text encoders have rapidly advanced the state of the art on many NLP tasks. We focus on one such model, BERT, and aim to quantify where linguistic information is captured within the network. We find that the model represents the steps of the traditional NLP pipeline in an interpretable and localizable way, and that the regions responsible for each step appear in the expected sequence: POS tagging, parsing, NER, semantic roles, then coreference. Qualitative analysis reveals that the model can and often does adjust this pipeline dynamically, revising lower-level decisions on the basis of disambiguating information from higher-level representations.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 29 Pith papers

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

  1. Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 small

    cs.LG 2022-11 conditional novelty 8.0

    GPT-2 small solves indirect object identification via a circuit of 26 attention heads organized into seven functional classes discovered through causal interventions.

  2. Demystifying Hidden-State Recurrence: Switchable Latent Reasoning with On-Policy Reinforcement Learning

    cs.LG 2026-06 unverdicted novelty 7.0

    SWITCH uses explicit <swi> and </swi> boundary tokens to make latent chain-of-thought compatible with on-policy RL (GRPO) and open to causal mechanistic probing, outperforming prior hidden-state recurrence methods.

  3. Toward Calibrated, Fair, and accurate Deepfake Detection

    cs.LG 2026-06 unverdicted novelty 7.0

    Face-Feature Tuning is a label-free logit remapping method that reduces FPR/TPR gaps across groups in deepfake detection while preserving overall accuracy.

  4. Parameter-Efficient Fine-Tuning with Learnable Rank

    cs.CL 2026-06 unverdicted novelty 7.0

    LR-LoRA learns per-layer adapter ranks during training and reports outperforming fixed-rank LoRA and other PEFT baselines on language understanding and commonsense reasoning tasks.

  5. Mechanistic Interpretability of ASR models using Sparse Autoencoders

    cs.CL 2026-05 unverdicted novelty 7.0

    Sparse autoencoders applied to Whisper ASR reveal monosemantic features across linguistic boundaries and demonstrate cross-lingual feature steering.

  6. Latent Space Probing for Adult Content Detection in Video Generative Models

    cs.CV 2026-04 unverdicted novelty 7.0

    Latent space probing on CogVideoX achieves 97.29% F1 for adult content detection on a new 11k-clip dataset with 4-6ms overhead.

  7. Computational Lesions in Multilingual Language Models Separate Shared and Language-specific Brain Alignment

    cs.CL 2026-04 unverdicted novelty 7.0

    Lesioning a shared core in multilingual LLMs drops whole-brain fMRI encoding correlation by 60.32%, while language-specific lesions selectively weaken predictions only for the matched native language.

  8. The Long Delay to Arithmetic Generalization: When Learned Representations Outrun Behavior

    cs.LG 2026-03 unverdicted novelty 7.0

    The grokking delay in encoder-decoder models on one-step Collatz prediction stems from decoder inability to use early-learned encoder representations of parity and residue structure, with numeral base acting as a stro...

  9. Neuroprobe: Evaluating Intracranial Brain Responses to Naturalistic Stimuli

    cs.LG 2025-09 accept novelty 7.0

    Neuroprobe is a new suite of decoding tasks on the BrainTreebank iEEG dataset for evaluating multi-modal language processing in the brain during naturalistic movie viewing.

  10. Massive Activations in Large Language Models

    cs.CL 2024-02 unverdicted novelty 7.0

    Massive activations are constant large values in LLMs that function as indispensable bias terms and concentrate attention probabilities on specific tokens.

  11. OPT: Open Pre-trained Transformer Language Models

    cs.CL 2022-05 unverdicted novelty 7.0

    OPT releases open decoder-only transformers up to 175B parameters that match GPT-3 performance at one-seventh the carbon cost, along with code and training logs.

  12. Graph-Native Reinforcement Learning Enables Traceable Scientific Hypothesis Generation through Conceptual Recombination

    cs.AI 2026-07 unverdicted novelty 6.0

    Graph-PRefLexOR fine-tunes graph-native models with GRPO to organize reasoning into phases, yielding 40-65% gains in traceable hypothesis generation and 2-3x semantic diversity on 100 materials science questions.

  13. Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models

    cs.CL 2026-06 unverdicted novelty 6.0

    LLMs recover dominant binomial orders from corpora but align less closely with exact preference distributions, with preference strength partially encoded in middle-to-late layers and manipulable via steering.

  14. A Systematic Study of Behavioral Cloning for Scientific Data Annotation

    cs.HC 2026-05 unverdicted novelty 6.0

    Introduces 9 synthetic annotation tasks and benchmarks for behavioral cloning, finding hierarchical skill learning, scaling benefits, effective multi-task pretraining, and shared internal representations of task phase...

  15. Ensemble Monitoring for AI Control: Diverse Signals Outweigh More Compute

    cs.AI 2026-05 unverdicted novelty 6.0

    Diverse ensembles of prompted and fine-tuned GPT-4.1-Mini monitors achieve 2.4x better detection of flawed code solutions than homogeneous ensembles on adversarial inputs.

  16. Monitoring Neural Training with Topology: A Footprint-Predictable Collapse Index

    cs.LG 2026-04 unverdicted novelty 6.0

    A composite Collapse Index based on incremental discrete Morse homology provides low-latency early warning of representational collapse during neural network training.

  17. LLM Safety From Within: Detecting Harmful Content with Internal Representations

    cs.AI 2026-04 unverdicted novelty 6.0

    SIREN identifies safety neurons via linear probing on internal LLM layers and combines them with adaptive weighting to detect harm, outperforming prior guard models with 250x fewer parameters.

  18. AIM: Asymmetric Information Masking for Visual Question Answering Continual Learning

    cs.CV 2026-04 unverdicted novelty 6.0

    AIM applies modality-specific masks to balance stability and plasticity in asymmetric VLMs, achieving SOTA average performance and reduced forgetting on continual VQA v2 and GQA while preserving generalization to nove...

  19. Geometric Routing Enables Causal Expert Control in Mixture of Experts

    cs.AI 2026-04 unverdicted novelty 6.0

    Cosine-similarity routing in low-dimensional space makes MoE experts monosemantic by construction and enables direct causal control via centroid interventions.

  20. A Layer-wise Analysis of Supervised Fine-Tuning

    cs.LG 2026-04 unverdicted novelty 6.0

    Middle layers (20-80%) remain stable during SFT while final layers are sensitive, enabling Mid-Block Efficient Tuning that outperforms LoRA by up to 10.2% on GSM8K with reduced parameter count.

  21. How Do Language Models Compose Functions?

    cs.CL 2025-10 conditional novelty 6.0

    LLMs solve compositional factual recall either by computing intermediates or directly, with mechanism choice correlated to translation geometry in embedding spaces.

  22. AIM-CoT: Active Information-driven Multimodal Chain-of-Thought for Vision-Language Reasoning

    cs.CV 2025-09 unverdicted novelty 6.0

    AIM-CoT enhances interleaved multimodal chain-of-thought reasoning by adding context-enhanced attention generation, active visual probing via information foraging, and dynamic attention-shift triggering.

  23. VisualBERT: A Simple and Performant Baseline for Vision and Language

    cs.CV 2019-08 conditional novelty 6.0

    VisualBERT is a Transformer model that implicitly aligns text and image regions through self-attention and achieves competitive or superior results on VQA, VCR, NLVR2, and Flickr30K after pre-training on captions.

  24. When Meaning Travels: A Granular Lens on Hybrid-MoE's Role in Idiomatic Understanding for Language Models

    cs.CL 2026-06 unverdicted novelty 5.0

    HybridMoE with controlled hybridization and idiomatic property signals yields 5-6% gains in figurative language representation for multilingual vision-language models.

  25. TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability

    cs.LG 2026-05 unverdicted novelty 5.0

    Task-aware pruning improves OOD model performance by realigning distorted OOD layerwise norm and pairwise-distance profiles with the task-adapted geometry observed on ID inputs.

  26. TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability

    cs.LG 2026-05 unverdicted novelty 5.0

    Task-aware pruning improves OOD performance by removing layers that distort task-adapted representation profiles, realigning OOD inputs with the geometry observed on ID data.

  27. HyperLens: Quantifying Cognitive Effort in LLMs with Fine-grained Confidence Trajectory

    cs.AI 2026-05 unverdicted novelty 5.0

    HyperLens reveals that deeper transformer layers magnify small confidence changes into fine-grained trajectories, allowing quantification of cognitive effort where complex tasks demand more and standard SFT can reduce it.

  28. Evaluating Document-Tuned Transformer Representations for Person-level Mental Health Assessment

    cs.CL 2026-06 unverdicted novelty 4.0

    Document-tuned transformers outperform base transformers by 13.4% Pearson r on person-level mental health prediction across two datasets and remain more accurate under text perturbations.

  29. FedMTFI: Feature Importance Based Optimized Multi Teacher Knowledge Distillation in Heterogeneous Federated Learning Environment

    cs.LG 2026-06 unverdicted novelty 4.0

    FedMTFI clusters heterogeneous clients, trains cluster prototypes, and applies multi-teacher distillation with SHAP to improve accuracy over standard FL in non-IID settings.