ResearchTrend.AI
Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens

Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens

Kaya Stechly
Karthik Valmeekam
Atharva Gundawar
Vardhan Palod
Subbarao Kambhampati
ArXivPDFHTML

Papers citing "Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens"

2 / 2 papers shown
Title
Neither Stochastic Parroting nor AGI: LLMs Solve Tasks through Context-Directed Extrapolation from Training Data Priors
Neither Stochastic Parroting nor AGI: LLMs Solve Tasks through Context-Directed Extrapolation from Training Data Priors
Harish Tayyar Madabushi
Melissa Torgbi
C. Bonial
33
0
0
29 May 2025
When Models Reason in Your Language: Controlling Thinking Trace Language Comes at the Cost of Accuracy
When Models Reason in Your Language: Controlling Thinking Trace Language Comes at the Cost of Accuracy
Jirui Qi
Shan Chen
Zidi Xiong
Raquel Fernández
Danielle S. Bitterman
Arianna Bisazza
LRM
24
0
0
28 May 2025