SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2301.13379 · 2023

Faithful Chain-of-Thought Reasoning

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 9 functions out of this paper's own repositories and ran 8 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
veronica320/faithful-cot canonical 8 of 9
FunctionStatusWhere it lives
extract_gold_answer Ran veronica320/faithful-cot/source/dataset/utils.py
code served (permissive licence) · get_code("f10320e98057847d")
is_correct Ran veronica320/faithful-cot/source/evaluate/evaluate_answer_acc.py
code served (permissive licence) · get_code("26a0c594eb40f371")
lcm Ran veronica320/faithful-cot/source/scratch.py
code served (permissive licence) · get_code("2410baeb6a8ae747")
load_data Ran veronica320/faithful-cot/source/dataset/utils.py
code served (permissive licence) · get_code("1298973dba46c252")
solve_it Ran veronica320/faithful-cot/source/scratch.py
code served (permissive licence) · get_code("a495f91be279d500")
solve_mwp Ran veronica320/faithful-cot/source/model/solver/MWP/math_solver.py
code served (permissive licence) · get_code("913ed052cb272f34")
str2num Ran veronica320/faithful-cot/source/dataset/utils.py
code served (permissive licence) · get_code("6fd802abadc37bc8")
timeout Ran veronica320/faithful-cot/source/model/codex.py
code served (permissive licence) · get_code("cf46a2fecbc874d2")
solve Not yet run veronica320/faithful-cot/source/model/solver/CLUTRR/CLUTRR_solver.py
code served (permissive licence) · get_code("09e91d6ca5d39a6d")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

While Chain-of-Thought (CoT) prompting boosts Language Models' (LM) performance on a gamut of complex reasoning tasks, the generated reasoning chain does not necessarily reflect how the model arrives at the answer (aka. faithfulness). We propose Faithful CoT, a reasoning framework involving two stages: Translation (Natural Language query $\rightarrow$ symbolic reasoning chain) and Problem Solving (reasoning chain $\rightarrow$ answer), using an LM and a deterministic solver respectively. This guarantees that the reasoning chain provides a faithful explanation of the final answer. Aside from interpretability, Faithful CoT also improves empirical performance: it outperforms standard CoT on 9 of 10 benchmarks from 4 diverse domains, with a relative accuracy gain of 6.3% on Math Word Problems (MWP), 3.4% on Planning, 5.5% on Multi-hop Question Answering (QA), and 21.4% on Relational Inference. Furthermore, with GPT-4 and Codex, it sets the new state-of-the-art few-shot performance on 7 datasets (with 95.0+ accuracy on 6 of them), showing a strong synergy between faithfulness and accuracy.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2301.13379")
get_code_for_paper("2301.13379")
have("2301.13379")

Connect an agent — have() is free.