| Challenge: | Instruction-tuned large language models employ structured templates to enforce format consistency during inference. |
| Approach: | They fine-tune instruction-tuning large language models with structured templates and evaluate their results across three axes: downstream task performance, alignment behavior, and output diversity. |
| Outcome: | The proposed model generates semantically similar outputs even under high temperature sampling and structural tokens in templates significantly constrain the model’s output space. |
Similar Papers
Scaling Data Diversity for Fine-Tuning Language Models in Human Alignment (2024.lrec-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) can reveal toxic or offensive content inadvertently or intentionally. |
| Approach: | They propose to control the diversity of both sides according to the number of samples for fine-tuning, which can directly reflect their impact. |
| Outcome: | The proposed approach improves the performance of large language models after fine-tuning. |
LLMs Are Biased Towards Output Formats! Systematically Evaluating and Mitigating Output Format Bias of LLMs (2025.naacl-long)
Copied to clipboard
Do Xuan Long, Ngoc-Hai Nguyen, Tiviatis Sim, Hieu Dao, Shafiq Joty, Kenji Kawaguchi, Nancy F. Chen, Min-Yen Kan
| Challenge: | Using format-following capabilities, state-of-the-art large language models (LLMs) can be leveraged to tailor outputs to specific task formats. |
| Approach: | They propose to define a format bias evaluation metric and establish effective strategies to reduce it. |
| Outcome: | The proposed evaluation reduces the variance in ChatGPT’s performance among wrapping formats from 235.33 to 0.71 (%2) |
Towards LLMs Robustness to Changes in Prompt Format Styles (2025.naacl-srw)
Copied to clipboard
| Challenge: | Existing prompt engineering techniques for adapting pre-trained LLMs to specific tasks are zero-shot prompting and few-shot supervised fine-tuning. |
| Approach: | They propose a technique for addressing prompt brittleness by diversifying the styles used in the prompt few-shot examples by using computer vision techniques. |
| Outcome: | Empirical results show that the proposed technique reduces style-induced prompt brittleness while improving overall performance across prompt variations and different datasets. |
Synthetic Eggs in Many Baskets: The Impact of Synthetic Data Diversity on LLM Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Increasing demand for training data is causing language models to be trained on synthetic data, a new study finds . fine-tuning models on synthetic datasets reduces self-preference bias . |
| Approach: | They investigate the impact of diversity of synthetic data on fine-tuned large language models. |
| Outcome: | The proposed model can mitigate distribution collapse, maintain diversity of output distribution, and reduce self-preference bias. |
Persona-Augmented Benchmarking: Evaluating LLMs Across Diverse Writing Styles (2025.emnlp-main)
Copied to clipboard
| Challenge: | Current benchmarks for evaluating Large Language Models do not capture the rich variety of communication patterns exhibited by humans. |
| Approach: | They propose a low-cost method to emulate diverse writing styles by rewriting evaluation prompts using persona-based LLM prompting. |
| Outcome: | The proposed method improves the external validity of the benchmarks for Large Language Models (LLMs) based on persona-based prompting. |
From Distributional to Overton Pluralism: Investigating Large Language Model Alignment (2025.naacl-long)
Copied to clipboard
| Challenge: | a large language model's (LLM) output distribution is changed by an alignment process . a recent study shows that aligned models surface information that cannot be recovered from base models without fine-tuning. |
| Approach: | They analyze two aspects of the alignment process that change output distributions . they find alignment suppresses irrelevant and unhelpful content . |
| Outcome: | The proposed model can be imitated without fine-tuning by using in-context examples and lower-resolution semantic hints about response content. |
Quantifying the Impact of Structured Output Format on Large Language Models through Causal Inference (2026.findings-eacl)
Copied to clipboard
| Challenge: | Prior studies have examined the impact of structured output on LLMs’ generation quality, often presenting one-way findings. |
| Approach: | They propose to derive five potential causal structures characterizing the influence of structured output on LLMs’ generation using one assumed and two guaranteed constraints. |
| Outcome: | The proposed pipeline can be extended to other modules and is not limited to structured output but can be used in industrial applications. |
Balancing Diversity and Risk in LLM Sampling: How to Select Your Method and Parameter for Open-Ended Text Generation (2025.acl-long)
Copied to clipboard
| Challenge: | Recent studies suggest that sampling-based decoding strategies can be used to optimize the output of Large Language Models (LLMs) . previous studies have shown that likelihood-maximization produces degenerate text which contains repetitive loops and incoherent context, especially in open-ended tasks. |
| Approach: | They propose to use a prefix tree to estimate the intrinsic capacity of a truncation sampling method by considering the trade-off between diversity and risk at each decoding step. |
| Outcome: | The proposed method is based on a prefix tree which preserves the context of a full sentence. |
More Aligned, Less Diverse? Analyzing the Grammar and Lexicon of Two Generations of LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | a growing number of studies compare LLMs with human-authored text . diversity is unclear, but it is important to understand what makes human and machine writing distinct . |
| Approach: | They compare syntactic properties of AI-generated and human-authored English news texts . they use the Head-Driven Phrase Structure Grammar and the English Resource Grammar . |
| Outcome: | The proposed model differs from human-authored English news text in two years. |
Understanding and Improving Information Preservation in Prompt Compression for LLMs (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Recent advances in large language models have enabled their successful application to a broad range of tasks. |
| Approach: | They propose a framework that allows for in-depth analysis of prompt compression methods. |
| Outcome: | The proposed framework analyzes state-of-the-art soft and hard compression methods . it shows that some fail to preserve key details from the original prompt, limiting performance on complex tasks. |