Recent advances in Large Language Models (LLMs) - particularly model scaling and test-time techniques - have greatly enhanced the reasoning capabilities of language models at the expense of higher inference costs. To lower inference costs, prior works train router models or deferral mechanisms that allocate easy queries to a small, efficient model, while forwarding harder queries to larger, more expensive models. However, these trained router models often lack robustness under domain shifts and require expensive data synthesis techniques such as Monte Carlo rollouts to obtain sufficient ground-truth routing labels for training. In this work, we propose Confidence-Guided Stepwise Model Routing for Cost-Efficient Reasoning (STEER), a domain-agnostic, framework that performs fine-grained, step-level routing between smaller and larger LLMs without utilizing external models. STEER leverages confidence scores from the smaller model’s logits prior to generating a reasoning step, so that the large model is invoked only when necessary. Extensive evaluations using different LLMs on a diverse set of challenging benchmarks across multiple domains such as Mathematical Reasoning, Multi-Hop QA, and Planning tasks indicate that STEER achieves competitive or enhanced accuracy while reducing inference costs (up to +20% accuracy with 48% less FLOPs compared to solely using the larger model on AIME), outperforming baselines that rely on trained external modules. Our results establish model-internal confidence as a robust, domain-agnostic signal for model routing, offering a scalable pathway for efficient LLM deployment.

Story visualization requires generating a coherent sequence of images that collectively form a narrative, yet existing evaluation metrics and datasets often overlook visual continuity and narrative diversity.In this paper, we introduce the Visual Context-Aware Metric for Story Visualization (VCMS), which uses large vision-language models to jointly assess caption fidelity and inter-image consistency, achieving Spearman's correlation comparable to human agreement on two benchmarks.Also, to address the shortcomings of narrowly defined datasets with low diversity, we propose a diffusion-augmented evaluation pipeline that blends diverse and controlled narrative elements at adjustable ratios, enabling the creation of evaluation sets tailored to specific objectives.By combining VCMS with this pipeline, we provide a scalable, human-aligned framework for evaluating story visualization models.

Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks, but their potential misuse for harmful purposes remains a significant concern.To strengthen defenses against such vulnerabilities, it is essential to investigate universal jailbreak attacks that exploit intrinsic weaknesses in the architecture and learning paradigms of LLMs.In response, we propose Harmful Prompt Laundering (HaPLa), a novel and broadly applicable jailbreaking technique that requires only black-box access to target models.HaPLa incorporates two primary strategies: 1) abductive framing, which instructs LLMs to infer plausible intermediate steps toward harmful activities, rather than directly responding to explicit harmful queries; and 2) symbolic encoding, a lightweight and flexible approach designed to obfuscate harmful content, given that current LLMs remain sensitive primarily to explicit harmful keywords.Experimental results show that HaPLa achieves over 95% attack success rate on GPT-series models and 70% across all targets.Further analysis with diverse symbolic encoding rules also reveals a fundamental challenge: it remains difficult to safely tune LLMs without significantly diminishing their helpfulness in responding to benign queries.

Although auto-regressive models excel in natural language processing, they often struggle to generate diverse text and provide limited controllability.Non-auto-regressive methods could be an alternative but often produce degenerate outputs and exhibit shortcomings in conditional generation.To address these challenges, we propose Diffusion-EAGS, a novel framework that integrates conditional masked language models into diffusion language models through the theoretical lens of a conditional Markov Random Field.In doing so, we propose entropy-adaptive Gibbs sampling and entropybased noise scheduling to counterbalance each model's shortcomings.Experimental results show that Diffusion-EAGS outperforms baselines and achieves the best quality-diversity tradeoff, demonstrating its effectiveness in nonautoregressive text generation.Context Jake was playing with his toys.He accidentally broke his favorite one.He cried a lot over it.His parents decided to replace it for him.Keyword not stop Jake just could not stop crying.Jake feel It made Jake feel So much better.would enjoy Jake said he would enjoy the new toy Context Neil was in Sofia, Bulgaria.He was enjoying a trip backpacking through Europe....He thought the food and culture in

Kang-il Lee, Hyukhun Koh, Dongryeol Lee, Seunghyun Yoon, Minsung Kim, Kyomin Jung. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

We introduce transductive program synthesis, a new formulation of the program synthesis task that explicitly leverages test inputs during synthesis. While prior approaches to program synthesis--whether based on natural language descriptions or input-output examples--typically aim to generalize from training examples, they often struggle with robustness, especially in real-world settings where training examples are limited and test inputs involve various edge cases. To address this, we propose a novel framework that improves robustness by treating synthesis as an active learning over a finite hypothesis class defined by programs' outputs. We use an LLM to predict outputs for selected test inputs and eliminate inconsistent hypotheses, where the inputs are chosen via a greedy maximin algorithm to minimize the number of LLM queries required. We evaluate our approach on four benchmarks: Playgol, MBPP+, 1D-ARC, and programmatic world modeling on MiniGrid. We demonstrate that our method significantly improves program synthesis in both accuracy and efficiency. We release our code at https://github.com/klee972/SYNTRA.

Minwoo Lee, Hyukhun Koh, Minsung Kim, Kyomin Jung. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Gender bias is a significant issue in machine translation, leading to ongoing research efforts in developing bias mitigation techniques.However, most works focus on debiasing bilingual models without much consideration for multilingual systems.In this paper, we specifically target the gender bias issue of multilingual machine translation models for unambiguous cases where there is a single correct translation, and propose a bias mitigation method based on a novel approach.Specifically, we propose Gender-Aware Contrastive Learning, GACL, which encodes contextual gender information into the representations of non-explicit gender words.Our method is target languageagnostic and is applicable to pre-trained multilingual machine translation models via finetuning.Through multilingual evaluation, we show that our approach improves gender accuracy by a wide margin without hampering translation performance.We also observe that incorporated gender information transfers and benefits other target languages regarding gender accuracy.Finally, we demonstrate that our method is applicable and beneficial to models of various sizes. 1

With the rapid advancement in deep generative models, recent neural Text-To-Speech (TTS) models have succeeded in synthesizing humanlike speech.There have been some efforts to generate speech with various prosody beyond monotonous prosody patterns.However, previous works have several limitations.First, typical TTS models depend on the scaled sampling temperature for boosting the diversity of prosody.Speech samples generated at high sampling temperatures often lack perceptual prosodic diversity, thereby hampering the naturalness of the speech.Second, the diversity among samples is neglected since the sampling procedure often focuses on a single speech sample rather than multiple ones.In this paper, we propose DPP-TTS: a text-to-speech model based on Determinantal Point Processes (DPPs) with a new objective function and prosody diversifying module.Our TTS model is capable of generating speech samples that simultaneously consider perceptual diversity in each sample and among multiple samples.We demonstrate that DPP-TTS generates speech samples with more diversified prosody than baselines in the side-by-side comparison test considering the naturalness of speech at the same time.