TROPT: An Open Framework for Unifying and Advancing Discrete Text Optimization

Hugging Face Daily Papers Papers
discrete-optimization text-optimization open-source framework red-teaming jailbreaks interpretability

Summary

TROPT is an open-source framework that unifies discrete text-trigger optimization, standardizing development and execution across domains like LLM jailbreaking and model interpretability. It includes over 15 optimizers and 30 recipes, lowering barriers for adoption and advancement.

Discrete text-trigger optimization -- searching for text sequences that, when ingested by a model, steer it toward a specified objective -- underpins model red-teaming (e.g., LLM jailbreaks), as well as auditing and interpretability. However, the current state of discrete optimizers hinders their adoption and progress. First, existing optimizers, when open-sourced at all, are scattered across research codebases tied to specific models, objectives, and problem domains. Second, optimizer variants proliferate, each requiring engineering overhead to use or extend, and remaining hard to compare head-to-head. Together, these raise the bar for adopting optimizers in existing or new domains, and for advancing them via new strategies. We address these gaps with TROPT, the first open-source framework that unifies discrete optimizers' execution and standardizes their development under a single interface. TROPT makes it easy to customize end-to-end optimization recipes by swapping any component -- models, objectives, and optimizers -- extending its reach across domains and new applications. TROPT currently ships with 30+ optimization recipes -- covering applications such as jailbreaking and probing model internals -- built from 15+ optimizers (spanning white-box to black-box access) and 15+ losses, from foundational to state-of-the-art methods. Demonstrating its utility, we leverage TROPT in several studies: (i) controlled, large-scale experiments comparing and enhancing optimization strategies for LLM jailbreaks, revealing potent-yet-underadopted techniques; and (ii) porting optimizers from one domain (e.g., LLM jailbreak) to new domains (e.g., corpus-poisoning embedding model). In all, TROPT significantly lowers the barrier to adopting and advancing discrete text optimization.
Original Article
View Cached Full Text

Cached at: 06/23/26, 09:44 PM

Paper page - TROPT: An Open Framework for Unifying and Advancing Discrete Text Optimization

Source: https://huggingface.co/papers/2606.23496

Abstract

A unified open-source framework for discrete text-trigger optimization that standardizes the development and execution of optimization strategies across various domains and applications.

Discrete text-trigger optimization-- searching for text sequences that, when ingested by a model, steer it toward a specified objective -- underpinsmodel red-teaming(e.g.,LLM jailbreaks), as well asauditingandinterpretability. However, the current state of discrete optimizers hinders their adoption and progress. First, existing optimizers, when open-sourced at all, are scattered across research codebases tied to specific models, objectives, and problem domains. Second, optimizer variants proliferate, each requiring engineering overhead to use or extend, and remaining hard to compare head-to-head. Together, these raise the bar for adopting optimizers in existing or new domains, and for advancing them via new strategies. We address these gaps with TROPT, the first open-source framework that unifies discrete optimizers’ execution and standardizes their development under a single interface. TROPT makes it easy to customize end-to-endoptimization recipesby swapping any component -- models, objectives, and optimizers -- extending its reach across domains and new applications. TROPT currently ships with 30+optimization recipes-- covering applications such as jailbreaking and probing model internals -- built from 15+ optimizers (spanning white-box toblack-box access) and 15+ losses, from foundational to state-of-the-art methods. Demonstrating its utility, we leverage TROPT in several studies: (i) controlled, large-scale experiments comparing and enhancingoptimization strategiesforLLM jailbreaks, revealing potent-yet-underadopted techniques; and (ii) porting optimizers from one domain (e.g., LLM jailbreak) to new domains (e.g., corpus-poisoning embedding model). In all, TROPT significantly lowers the barrier to adopting and advancing discrete text optimization.

View arXiv pageView PDFProject pageGitHub5Add to collection

Get this paper in your agent:

hf papers read 2606\.23496

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.23496 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.23496 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.23496 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

optimize_anything: A Universal API for Optimizing any Text Parameter

arXiv cs.CL

This paper presents optimize_anything, a universal LLM-based optimization system for text artifacts that achieves state-of-the-art results across diverse tasks including agent architecture discovery, scheduling, CUDA kernel generation, and packing, demonstrating general-purpose text optimization.

OmniOpt: Taxonomy, Geometry, and Benchmarking of Modern Optimizers

Hugging Face Daily Papers

OmniOpt presents a unified framework for optimizer selection in large-scale model training, combining meta-pipeline transformations, norm-constrained linear minimization oracles, and a cross-domain benchmark to systematically analyze optimizer families and their trade-offs.

TTPO: Test-Time Policy Optimization

Hugging Face Daily Papers

TTPO introduces a label-free method for test-time training in mathematical reasoning that matches supervised performance by asymmetrically handling agreeing and disagreeing rollouts.