Tootfinder

A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
Jinyi Han, Xinyi Wang, Haiquan Zhao, Tingyun li, Zishang Jiang, Sihang Jiang, Jiaqing Liang, Xin Lin, Weikang Zhou, Zeye Sun, Fei Yu, Yanghua Xiao
https://arxiv.org/abs/2508.12903

A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models
Recent advances in self-refinement have demonstrated significant potential for improving the outputs of large language models (LLMs) through iterative refinement. However, most existing self-refinement methods rely on a reactive process with a fixed number of iterations, making it difficult to determine the optimal timing and content of refinement based on the evolving generation context. Inspired by the way humans dynamically refine their thoughts during execution, we propose ProActive Self-Re…

Refinement-Types Driven Development: A study
Facundo Dom\'inguez, Arnaud Spiwack
https://arxiv.org/abs/2509.15005 https://arxiv.org/pdf/2509.15005

Refinement-Types Driven Development: A study
This paper advocates for the broader application of SMT solvers in everyday programming, challenging the conventional wisdom that these tools are solely for formal methods and verification. We claim that SMT solvers, when seamlessly integrated into a compiler's static checks, significantly enhance the capabilities of ordinary type checkers in program composition. Specifically, we argue that refinement types, as embodied by Liquid Haskell, enable the use of SMT solvers in mundane programming tas…

Structural Abstraction and Refinement for Probabilistic Programs
Guanyan Li, Juanen Li, Zhilei Han, Peixin Wang, Hongfei Fu, Fei He
https://arxiv.org/abs/2508.12344 https://

Structural Abstraction and Refinement for Probabilistic Programs
In this paper, we present structural abstraction refinement, a novel framework for verifying the threshold problem of probabilistic programs. Our approach represents the structure of a Probabilistic Control-Flow Automaton (PCFA) as a Markov Decision Process (MDP) by abstracting away statement semantics. The maximum reachability of the MDP naturally provides a proper upper bound of the violation probability, termed the structural upper bound. This introduces a fresh ``structural'' characterizati…

Multi-Threaded Software Model Checking via Parallel Trace Abstraction Refinement
Max Barth, Marie-Christine Jakobs
https://arxiv.org/abs/2509.13699 https://

Multi-Threaded Software Model Checking via Parallel Trace Abstraction Refinement
Automatic software verification is a valuable means for software quality assurance. However, automatic verification and in particular software model checking can be time-consuming, which hinders their practical applicability e.g., the use in continuous integration. One solution to address the issue is to reduce the response time of the verification procedure by leveraging today's multi-core CPUs. In this paper, we propose a solution to parallelize trace abstraction, an abstraction-based appro…

Listening, Imagining \& Refining: A Heuristic Optimized ASR Correction Framework with LLMs
Yutong Liu, Ziyue Zhang, Yongbin Yu, Xiangxiang Wang, Yuqing Cai, Nyima Tashi
https://arxiv.org/abs/2509.15095

Listening, Imagining \& Refining: A Heuristic Optimized ASR Correction Framework with LLMs
Automatic Speech Recognition (ASR) systems remain prone to errors that affect downstream applications. In this paper, we propose LIR-ASR, a heuristic optimized iterative correction framework using LLMs, inspired by human auditory perception. LIR-ASR applies a "Listening-Imagining-Refining" strategy, generating phonetic variants and refining them in context. A heuristic optimization with finite state machine (FSM) is introduced to prevent the correction process from being trapped in local optima…

‘The dungeon’ at Louisiana’s notorious prison reopens as Ice detention center
The Louisiana state penitentiary at Angola – the largest maximum security prison in the country – was out of bed space for “violent offenders” who would be “transferred to its facilities”,
the governor warned in an executive order.
The emergency declaration allowed for the rapid refurbishing of a notorious, shuttered housing unit at Angola formerly known as Camp J
– commonly referred to by p…

Yes, even in Manchester.
Blackberrying from July, early grapes and huge crop in what was a marginal location, apples ripe at least a month early.
Early ripening of berries in UK shows nature is under stress, say experts | UK weather | The Guardian
https://archive.ph/xiNn5

Refining the deep sub-barrier 12C 12C excitation function with STELLA
J. Nippert, S. Courtin, M. Heine, D. G. Jenkins, P. Adsley, A. Bonhomme, R. Canavan, D. Curien, T. Dumont, E. Gregor, G. Harmant, E. Monpribat, L. Morrison, M. Moukaddam, M. Richer, M. Rudigier, J. G. Vega Romero, W. N. Catford, P. Cotte, S. Della Negra, G. Haefner, F. Hammache, J. Lesrel, S. Pascu, Zs. Podoly\'ak, P. H. Regan, I. Ribaud, N. de S\'er\'eville, C. Stodel, J. Vesi\'c

Refining the deep sub-barrier 12C+12C excitation function with STELLA
Purpose: We have investigated the cross section around the lowest direct coincident gamma-particle measurements, where previously only limits could be established with the aim of obtaining a detailed description of the excitation function. We have furthermore analysed the ratio of extreme decay branching into the first excited state of daughter nuclei with alpha or proton emission at relative kinetic energies where previous measurements are in disagreement with each other. Conclusions: Our find…

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement
Ali Salamatian, Amirhossein Abaskohi, Wan-Cyuan Fan, Mir Rayat Imtiaz Hossain, Leonid Sigal, Giuseppe Carenini
https://arxiv.org/abs/2509.13282

ChartGaze: Enhancing Chart Understanding in LVLMs with Eye-Tracking Guided Attention Refinement
Charts are a crucial visual medium for communicating and representing information. While Large Vision-Language Models (LVLMs) have made progress on chart question answering (CQA), the task remains challenging, particularly when models attend to irrelevant regions of the chart. In this work, we present ChartGaze, a new eye-tracking dataset that captures human gaze patterns during chart reasoning tasks. Through a systematic comparison of human and model attention, we find that LVLMs often diverge…

Replaced article(s) found for cs.CL. https://arxiv.org/list/cs.CL/new
[3/3]:
- Audited Reasoning Refinement: Fine-Tuning Language Models via LLM-Guided Step-Wise Evaluation and...
Sumanta Bhattacharyya, Sara Riazi, Pedram Rooshenas