arXiv AI By Nathan Sheffield, Alek Westover, Lukas Finnveden, Alexa Pan, Julian Stastny, Ryan Greenblatt

An Operationalization of Opaque Serial Depth

Read the original on arXiv AI →

The paper introduces a new metric, NLS depth, to quantify the opaque serial depth of AI models—essentially the longest computation path that bypasses interpretable bottlenecks. It defines natural‑language‑rooted nodes as the bottlenecks and shows that NLS depth can be determined from architecture and training settings alone, remaining unaffected by post‑training techniques. Across open‑weight models, NLS depth correlates strongly with layer count and has increased slowly, suggesting that future scaling of standard transformers will only modestly raise it, while latent‑reasoning architectures could raise it substantially.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Computation and Language
Oct 21

From Abstraction to Action: DSLs for Natural Language Abstraction Interpretation and Execution

The paper introduces domain-specific languages (DSLs) as an explicit representation layer to bridge the gap between high-level natural-language abstractions and concrete actions. Experiments on the visually-grounded “Hexagons” task and the “Text2CAD” benchmark show that code generation using a human-curated DSL yields dramatic performance gains, especially at higher abstraction levels, while automatically induced DSLs only partially recover these gains. The study highlights that the right representation can close much of the performance gap caused by abstraction, but that automatically learning such representations remains a challenge.

By Amit Lisha, Reut Tsarfaty
arXiv AI
Jul 10

Rethinking LLM-as-a-Judge: Representation-as-a-Judge with Small Language Models via Semantic Capacity Asymmetry

arXiv:2601. 22588v2 Announce Type: replace-cross Abstract: Large language models (LLMs) are widely used as reference-free evaluators via prompting, but this "LLM-as-a-Judge" paradigm is costly, opaque, and sensitive to prompt design.

By Zhuochun Li, Yong Zhang, Ming Li, Yuelyu Ji, Yiming Zeng, Ning Cheng, Yun Zhu, Yanmeng Wang, Shaojun Wang, Jing Xiao, Daqing He