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.
The conceptarc benchmark: Evaluating under- standing and generalization in the arc domain
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
verdicts
UNVERDICTED 4representative citing papers
VisAnalog is a new controlled benchmark showing VLMs substantially underperform humans on visual concept transfer under one- to four-step deterministic transformations, with relation inference as the main failure mode.
TRM with 7M parameters achieves 45% accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, surpassing most LLMs with under 0.01% of their parameters.
RefusalGuard constrains updates in hidden representation space to preserve safety-relevant geometric structure during fine-tuning, maintaining low attack success rates on safety benchmarks while preserving task performance.
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.
-
Less is More: Recursive Reasoning with Tiny Networks
TRM with 7M parameters achieves 45% accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, surpassing most LLMs with under 0.01% of their parameters.
-
RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs
RefusalGuard constrains updates in hidden representation space to preserve safety-relevant geometric structure during fine-tuning, maintaining low attack success rates on safety benchmarks while preserving task performance.