Hugging Face Trending Papers July 13, 2026

Amplitude-Only FFN Intervention for Tool-Structured LLM Inference Method: Gated Evaluation Protocol, and Cross-Model Empirical Results

Read the original on Hugging Face Trending Papers →

Large language models increasingly operate as tool-using agents, where small format, argument, or function-call errors can invalidate otherwise plausible responses. We study inference-time feed-forward network (FFN) intervention for improving structured outputs without retraining model weights.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv Machine Learning

Aug 4

Amplitude-Only FFN Intervention for Tool-Structured LLM Inference Method: Gated Evaluation Protocol, and Cross-Model Empirical Results

arXiv:2607. 11183v2 Announce Type: cross Abstract: Large language models increasingly operate as tool-using agents, where small format, argument, or function-call errors can invalidate otherwise plausible responses.

By Sheng Xu, Junhua Wang, Boyuan Huang, Ke Jia, Jiadun Zhu, Zhen Chen

llms agents

arXiv AI

Jul 15

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

arXiv:2607. 12463v1 Announce Type: new Abstract: Coding agents must integrate external tool returns into ongoing reasoning - a capability that standard left-to-right pretraining on code exposes only in its forward direction.

By Yubo Wang, Jiarong Liang, Yuxuan Zhang, Xuye Liu, Cong Wei, Yuyu Zhang, Ping Nie, Wenhu Chen

agents benchmarks safety

arXiv AI

Jun 2

Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning

arXiv:2606. 01682v1 Announce Type: cross Abstract: Selecting the best response from multiple small-model samples using a stronger scorer is a simple inference-time strategy, but fails when the small model has already committed to incorrect reasoning paths.

By Atoosa Chegini, Soheil Feizi

llms reinforcement-learning benchmarks safety

arXiv AI

Jul 22

Operational Proto-Introspection in Looped Language Models: Process-Quality Taps, Executable Branching, and the Readout-Control Boundary

arXiv:2607. 18553v1 Announce Type: cross Abstract: Can a language model read the quality of ongoing computation, and can an external intervention turn that readout into better outcomes?

By Jan Kirin

llms fine-tuning

arXiv Machine Learning

Aug 3

OpsLLM: Construction of Large Language Model for Software Operations with Multi-stage Learning

arXiv:2605. 02906v3 Announce Type: replace Abstract: In the field of software operations, Large Language Models (LLMs) have attracted increasing attention.

By Jingkai He, Pengfei Chen, Chenghui Wu, Shuang Liang, Ye Li, Gou Tan, Xidao Wen, Chuanfu Zhang, Fang Situ, Qi Zhou

llms nlp reinforcement-learning fine-tuning

arXiv AI

Jul 9

Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?

arXiv:2605. 03195v2 Announce Type: replace Abstract: Modern coding agents increasingly delegate specialized subtasks to subagents, which are smaller, focused agentic loops that handle narrow responsibilities like search, debugging or terminal execution.

By Spandan Garg, Vikram Nitin, Yufan Huang

llms agents reinforcement-learning fine-tuning benchmarks

Related stories

Amplitude-Only FFN Intervention for Tool-Structured LLM Inference Method: Gated Evaluation Protocol, and Cross-Model Empirical Results

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning

Operational Proto-Introspection in Looped Language Models: Process-Quality Taps, Executable Branching, and the Readout-Control Boundary

OpsLLM: Construction of Large Language Model for Software Operations with Multi-stage Learning

Terminus-4B: Can a Smaller Model Replace Frontier LLMs at Agentic Execution Tasks?