Code for the paper "Implicit Representations of Meaning in Neural Language Models"

Last update: Nov 03, 2022

Related tags

Overview

Implicit Representations of Meaning in Neural Language Models

Preliminaries

Create and set up a conda environment as follows:

conda create -n state-probes python=3.7
conda activate state-probes
pip install -r requirements.txt

Install the appropriate torch 1.7.0 for your cuda version:

conda install pytorch==1.7.0 cudatoolkit=<cuda_version> -c pytorch

Before running any command below, run

export PYTHONPATH=.
export TOKENIZERS_PARALLELISM=true

Data

The Alchemy data is downloaded from their website.

wget https://nlp.stanford.edu/projects/scone/scone.zip
unzip scone.zip

The synthetic version of alchemy was generated by running:

echo 0 > id #the code requires a file called id with a number in it ...
python alchemy_artificial_generator.py --num_scenarios 3600 --output synth_alchemy_train
python alchemy_artificial_generator.py --num_scenarios 500 --output synth_alchemy_dev
python alchemy_artificial_generator.py --num_scenarios 900 --output synth_alchemy_test

You can also just download our generated data through:

wget http://web.mit.edu/bzl/www/synth_alchemy.tar.gz
tar -xzvf synth_alchemy.tar.gz

The Textworld data is under

wget http://web.mit.edu/bzl/www/tw_data.tar.gz
tar -xzvf tw_data.tar.gz

LM Training

To train a BART or T5 model on Alchemy data

python scripts/train_alchemy.py \
    --arch [t5|bart] [--no_pretrain] \
    [--synthetic] --encode_init_state NL

Saves model checkpoints under sconeModels/*.

To train a BART or T5 model on Textworld data

python scripts/train_textworld.py \
    --arch [t5/bart] [--no_pretrain] \
    --data tw_data/simple_traces --gamefile tw_games/simple_games

Saves model checkpoints nder twModels/*.

Probe Training & Evaluation

Alchemy

The main probe command is as follows:

python scripts/probe_alchemy.py \
    --arch [bart|t5] --lm_save_path <path_to_lm_checkpoint> [--no_pretrain] \
    --encode_init_state NL --nonsynthetic \
    --probe_target single_beaker_final.NL --localizer_type single_beaker_init_full \
    --probe_type linear --probe_agg_method avg \
    --encode_tgt_state NL.[bart|t5] --tgt_agg_method avg \
    --batchsize 128 --eval_batchsize 1024 --lr 1e-4

For evaluation, add --eval_only --probe_save_path <path_to_probe_checkpoint>. This will save model predictions to a .jsonl file under the same directory as the probe checkpoint.

Add --control_input for No LM experiment.

Change --probe_target to single_beaker_init.NL to decode initial state.

For localization experiments, set --localizer_type single_beaker_init_{$i}.offset{$off} for some token i in {article, pos.[R0|R1|R2], beaker.[R0|R1], verb, amount, color, end_punct} and some integer offset off between 0 and 6.

Saves probe checkpoints under probe_models_alchemy/*.

Intervention experiment results follow from running the script:

python scripts/intervention.py \
    --arch [bart|t5] \
    --encode_init_state NL \
    --create_type drain_1 \
    --lm_save_path <path_to_lm_checkpoint>

which creates two contexts and replaces a select few encoded tokens to modify the underlying belief state.

Textworld

Begin by creating the full set of encoded proposition representations

python scripts/get_all_tw_facts.py \
    --data tw_data/simple_traces --gamefile tw_data/simple_games \
    --state_model_arch [bart|t5] \
    --probe_target belief_facts_pair \
    --state_model_path [None|pretrain|<path_to_lm_checkpoint>] \
    --out_file <path_to_prop_encodings>

Run the probe with

python scripts/probe_textworld.py \
    --arch [bart|t5] --data tw_data/simple_traces --gamefile tw_data/simple_games \
    --probe_target final.full_belief_facts_pair --encode_tgt_state NL.[bart|t5] \
    --localizer_type belief_facts_pair_[first|last|all] --probe_type 3linear_classify \
    --probe_agg_method avg --tgt_agg_method avg \
    --lm_save_path <path_to_lm_checkpoint> [--no_pretrain] \
    --ents_to_states_file <path_to_prop_encodings> \
    --eval_batchsize 256 --batchsize 32

For evaluation, add --eval_only --probe_save_path <path_to_probe_checkpoint>. This will save model predictions to a .jsonl file under the same directory as the probe checkpoint.

Add --control_input for No LM experiment.

Change --probe_target to init.full_belief_facts_pair to decode initial state.

For remap experiments, change --probe_target to final.full_belief_facts_pair.control_with_rooms.

For decoding from just one side of propositions, replace any instance of belief_facts_pair (in --probe_target and --localizer_type) with belief_facts_single and rerun both commands (first get the full proposition encodings, then run the probe).

Saves probe checkpoints under probe_models_textworld/*.

Print Metrics

Print full metrics (state EM, entity EM, subdivided by relations vs. propositions, etc.) using scripts/print_metrics.py.

python scripts/print_metrics.py \
    --arch [bart|t5] --domain [alchemy|textworld] \
    --pred_files <path_to_model_predictions_1>,<path_to_model_predictions_2>,<path_to_model_predictions_3>,... \
    [--use_remap_domain --remap_fn <path_to_remap_model_predictions>] \
    [--single_side_probe]

Code for the paper "Implicit Representations of Meaning in Neural Language Models"

Related tags

Overview

Implicit Representations of Meaning in Neural Language Models

Preliminaries

Data

LM Training

Probe Training & Evaluation

Alchemy

Textworld

Print Metrics

Owner

Belinda Li

This repository includes the official project for the paper: TransMix: Attend to Mix for Vision Transformers.

A Python library for generating new text from existing samples.

Gas detection for Raspberry Pi using ADS1x15 and MQ-2 sensors

CAMPARI: Camera-Aware Decomposed Generative Neural Radiance Fields

A micro-game "flappy bird".

Pytorch Implementation of Residual Vision Transformers(ResViT)

Codes for NeurIPS 2021 paper "On the Equivalence between Neural Network and Support Vector Machine".

A GOOD REPRESENTATION DETECTS NOISY LABELS

[cvpr22] Perturbed and Strict Mean Teachers for Semi-supervised Semantic Segmentation

The FIRST GANs-based omics-to-omics translation framework

TensorFlow implementation of Style Transfer Generative Adversarial Networks: Learning to Play Chess Differently.

10x faster matrix and vector operations

Multi-view 3D reconstruction using neural rendering. Unofficial implementation of UNISURF, VolSDF, NeuS and more.

code for Multi-scale Matching Networks for Semantic Correspondence, ICCV

BiSeNet based on pytorch

Video Swin Transformer - PyTorch

Evolving neural network parameters in JAX.

Code for our paper Domain Adaptive Semantic Segmentation with Self-Supervised Depth Estimation

Pipeline for employing a Lightweight deep learning models for LOW-power systems

PyTorch implementation of the wavelet analysis from Torrence & Compo