NEWTON: Are Large Language Models Capable of Physical Reasoning?
Bibliographic record
Abstract
Large Language Models (LLMs), through their contextualized representations, have been empirically proven to encapsulate syntactic, semantic, word sense, and common-sense knowledge.However, there has been limited exploration of their physical reasoning abilities, specifically concerning the crucial attributes for comprehending everyday objects.To address this gap, we introduce NEWTON, a repository and benchmark for evaluating the physics reasoning skills of LLMs.Further, to enable domain-specific adaptation of this benchmark, we present a pipeline to enable researchers to generate a variant of this benchmark that has been customized to the objects and attributes relevant for their application.The NEWTON repository comprises a collection of 2800 object-attribute pairs, providing the foundation for generating infinitescale assessment templates.The NEWTON benchmark consists of 160K QA questions, curated using the NEWTON repository to investigate the physical reasoning capabilities of several mainstream language models across foundational, explicit, and implicit reasoning tasks.Through extensive empirical analysis, our results highlight the capabilities of LLMs for physical reasoning.We find that LLMs like GPT-4 demonstrate strong reasoning capabilities in scenario-based tasks but exhibit less consistency in object-attribute reasoning compared to humans (50% vs. 84%).Furthermore, the NEWTON platform demonstrates its potential for evaluating and enhancing language models, paving the way for their integration into physically grounded settings, such as robotic manipulation.Project site: https: //newtonreasoning.github.ioTrack 3: Implicit Scenariobased Analysis Scenario 1 Attribute: Stiffness (high), Brittleness (low) Context: I am packing a backpack.Question: Which of , , , should I put at the bottom?Scenario 2 Attribute: Malleability (high), Elasticity (high) Context: I have an irregularly shaped space in my suitcase, and four objects with the same volume.Question: Which of , , , could fit in that space?Scenario 3 Attribute: Surface Hardness (high), Surface Smoothness (low) Context: I need an object to place sandpaper above.Question: Which of , , , is the most suitable?Scenario 4 Attribute: Sharpness (high) Context: I am trying to open some plastic packaging.Question: Which of , , , can help me open?Scenario 5 Attribute: Softness (high), Elasticity (high) Context: I am wrapping fragile gifts and want to protect them from impact.Question: Which of , , , should I choose for cushioning?Scenario 6 Attribute: Surface hardness (high), brittleness (low), stiffness (high) Context: I need to hammer a nail into a solid wooden board.Question: Which of , , , should I choose?Scenario 7 Attribute: Sharpness (high) Context: I need to prepare a work table as a play area for kids.Question: Which of , , , should I remove?Scenario 8 Attribute: Brittleness (low), Elasticity (high) Context: I have a robot which sorts objects by tossing them to bins.Question: Which of , , , should I remove?Scenario 9 Attribute: Softness (high), Elasticity (high) Context: I need an object to provide insulation from a sharp edge on a piece of furniture.Question: Which of , , , should I choose?
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.009 | 0.072 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.005 | 0.014 |
| Open science | 0.003 | 0.004 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.011 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".