NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
arXiv preprint arXiv:2305.19555 , year=
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
representative citing papers
DiARC improves LLM performance on ARC-like benchmarks by constructing and training on preference pairs from three types of negative samples while keeping demonstrations fixed.
BFS-based LLM framework reduces causal graph discovery queries from quadratic to linear while incorporating observational data and reporting state-of-the-art results on real graphs.
In the Flux environment, RL agents with explicit latent state access achieve ~79% win rate versus ~11% for LLMs on long-horizon tasks, illustrating limitations of sequence prediction for dynamic reasoning.
citing papers explorer
-
Gradient-Based Program Synthesis with Neurally Interpreted Languages
NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
-
DiARC: Distinguishing Positive and Negative Samples Helps Improving ARC-like Reasoning Ability of Large Language Models
DiARC improves LLM performance on ARC-like benchmarks by constructing and training on preference pairs from three types of negative samples while keeping demonstrations fixed.
-
Efficient Causal Graph Discovery Using Large Language Models
BFS-based LLM framework reduces causal graph discovery queries from quadratic to linear while incorporating observational data and reporting state-of-the-art results on real graphs.
-
Why We Need World Models for AGI: Where LLMs Fail and How World Models May Outperform
In the Flux environment, RL agents with explicit latent state access achieve ~79% win rate versus ~11% for LLMs on long-horizon tasks, illustrating limitations of sequence prediction for dynamic reasoning.
- The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes