AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Leap-Of-Thought: Teaching Pre-Trained Models to Systematically Reason Over Implicit Knowledge

Alon Talmor, Oyvind Tafjord, Peter Clark, Yoav Goldberg, Jonathan Berant

NeurIPS 2020 · Large language models › Retrieval and knowledge · approach: Knowledge graph reasoning with LLMs · 27 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
fine-tune on generated hypernymy data and give no explicit hypernym rule (implicit reasoning) vs hypothesis onlyRoBERTa-Large · Hypernymy testaccuracy (%): 65.2 -> 88.8Improved
add RuleTaker+20Q-only training instead of hypernymy training (no direct implicit training) (ablation)RoBERTa-Large · Hypernymy test, Implicit Reasoningaccuracy (%): 65.4 -> 79.1 (vs 88.8 with direct training)Improved
replace RoBERTa-Large with ESIM + GloVe (ablation)hypernymy reasoning · Hypernymy Implicit Reasoningaccuracy (%): 88.8 -> 76.1Hurt
intervene by adding correct rules for the model's incorrect implicit beliefs (ablation)RoBERTa implicit reasoning · Hypernymy testaccuracy (%): 88.8 -> 98.3Improved
intervene by adding correct rules for incorrect beliefs (ablation)RoBERTa implicit reasoning · Imaginaryaccuracy (%): 76.9 -> 94.1Improved
add explicit member facts and counting training to hypothesis-onlyRoBERTa-Large · Counting (1,K-1) subsetaccuracy (%): 64.1 -> 73Improved
drop the quantity fact from test examples (ablation)RoBERTa counting model · Counting (1,K-1) subsetaccuracy (%): 73 -> 64.4Hurt
train on all four datasets instead of subsets (ablation)RoBERTa-Large · Multi-skill-setoverall accuracy (%): 73.5 -> 85.4Improved
add Counting data to RuleTaker+20Q training (ablation)RoBERTa-Large · Multi-skill-set counting skillaccuracy (%): 55.3 -> 83.0Improved
add member facts in ascending (least-known first) rather than descending model-probability order (ablation)RoBERTa counting model · Counting analysis, c=0.25accuracy: +5.5 points at c=0.25, +2.7 at c=0.75Improved
test with explicit knowledge present but no hypernym rule vs no context (distractors) (ablation)RoBERTa · Hypernymy Hypothesis-only vs Statement-only-no-contextaccuracy (%): 65.2 vs 66.7No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy