TEFM: Token-Efficient Faithful Modeling for Structured Data

TEFM is a new framework for applying LLMs to critical domains by reducing token consumption and maintaining faithfulness. Experiments across Qwen3, Gemma-2, and Phi-4 show significant token reduction with competitive accuracy and faithful rationales.

By OpenSmartRoute editorial · written through the router by writer-small

From arXiv cs.CL

TEFM addresses the challenges of applying large language models to domains requiring structured data analysis. The framework focuses on token efficiency through the use of Behavioral Code tokens, compressing lengthy observations. This approach dramatically reduces token consumption while minimizing information loss. TEFM also prioritizes faithfulness by employing a dual-fidelity objective. This objective simultaneously optimizes code-level reconstruction and prediction-level fidelity, identifying minimal sufficient feature subsets. Experiments across clinical and security domains demonstrate competitive classification accuracy with approximately 1% token retention and 2% token retention respectively. These results indicate a substantial reduction in resource requirements.

Source: https://arxiv.org/abs/2609.09552

Keep reading

Looped GPT-BERT Achieves Comparable Performance with Fewer Parameters

Researchers demonstrated Looped GPT-BERT, a model utilizing depth-wise parameter sharing and recurrent traversals, achieving performance comparable to larger GPT-2 and GPT-BERT models on benchmarks like BLiMP and GLUE. The design, trained on a 7.48M-word corpus, offers a more efficient approach to language modeling.

LLMsSep 10, 20261 min read

Algorithmic Speech: Three Mutations and Legal Implications

This research traces the evolution of algorithmic outputs, identifying three key mutations – speech as data, speech as engagement, and generative text – and their associated legal and social consequences. It provides a framework for understanding these changes and their impact on freedom of expression and informational privacy.

TEFM: Token-Efficient Faithful Modeling for Structured Data

Related posts

Looped GPT-BERT Achieves Comparable Performance with Fewer Parameters

Algorithmic Speech: Three Mutations and Legal Implications

RMS-RSP: Prioritizing Rationale Supervision in Medical QA