Papers by Tingchen Fu
There Are a Thousand Hamlets in a Thousand People’s Eyes: Enhancing Knowledge-grounded Dialogue with Personal Memory (2022.acl-long)
Copied to clipboard
| Challenge: | Existing methods for knowledge selection focus on relevance between knowledge and dialogue context, ignoring personal preference for knowledge. |
| Approach: | They propose to introduce personal memory into knowledge selection in chatbots to address personalization issue by integrating personal memory and inverse mapping into a closed loop. |
| Outcome: | The proposed method outperforms existing methods significantly on automatic evaluation and human evaluation. |
Unlocking Decoding-time Controllability: Gradient-Free Multi-Objective Alignment with Contrastive Prompts (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods for aligning large language models with human preferences are poor in extensibility and require significant retraining. |
| Approach: | They propose a multi-objective alignment approach that constructs an expert prompt and an adversarial prompt for each alignment objective to contrast at the decoding time. |
| Outcome: | The proposed approach is superior to existing methods in obtaining a well-distributed Pareto front among different alignment objectives. |
Logic Unveils Truth, While Disguise Obscures It: Transition Logic Augmented Response Selection for Multi-Turn Dialogue (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods of negative samples tend to yield false negatives due to one-to-many property in open-domain dialogue. |
| Approach: | They propose a sequential variational ladder auto-encoder to capture one-to-many transition pattern of multiple characteristics in open-domain dialogue. |
| Outcome: | The proposed approach improves the performance of a retrieval dialogue system on two benchmarks. |
BBA: Bi-Modal Behavioral Alignment for Reasoning with Large Vision-Language Models (2024.findings-acl)
Copied to clipboard
Xueliang Zhao, Xinting Huang, Tingchen Fu, Qintong Li, Shansan Gong, Lemao Liu, Wei Bi, Lingpeng Kong
| Challenge: | Multimodal reasoning is a key capability for large vision-language models . however, the vanilla Chain-of-Thought method fails to address critical steps in multi-step reasoning tasks. |
| Approach: | They propose a bi-modal Behavioral Alignment method to augment multimodal reasoning . they use domain-specific language to integrate multimodal information into a precise alternative form . |
| Outcome: | The proposed method significantly improves GPT-4V(ision) on geometry problem solving, chess positional advantage prediction and molecular property prediction. |
Towards Efficient Dialogue Pre-training with Transferable and Interpretable Latent Structure (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing models that use millions of parameters on massive data are inefficient and lack interpretability. |
| Approach: | They propose a model with a latent structure that is easily transferable from the general domain to downstream tasks in a lightweight and transparent way. |
| Outcome: | The proposed model performs better than four strong baseline models in terms of automatic and human evaluations and is 5x faster than the strongest baseline model. |
Same Question, Different Words: A Latent Adversarial Framework for Prompt Robustness (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing solutions to the problem of semantically-preserving variations of prompts are expensive and require trial-and-error prompt engineering. |
| Approach: | They propose a dual-loop adversarial framework that optimizes a trainable perturbation as "latent continuous paraphrase" they demonstrate a 0.5%-4% improvement on worst-case win-rate on the RobustAlpaca benchmark . |
| Outcome: | Extensive experiments show that the proposed framework improves on the RobustAlpaca benchmark with a 0.5%-4% improvement on the worst-case win-rate. |
Learning to Express in Knowledge-Grounded Conversation (2022.naacl-main)
Copied to clipboard
| Challenge: | Existing models focus on synthesizing a dialogue with proper knowledge, but neglect that the same knowledge could be expressed differently even under the same context. |
| Approach: | They propose a model that ground dialogue generation by extra knowledge by analyzing the structure of the response and the content style of each part. |
| Outcome: | The proposed model can learn the structure style defined by a few examples and generate responses in desired content style. |
Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in reasoning-oriented models have demonstrated impressive capabilities in mathematical reasoning, but their ability to adhere to user directives remains underexplored. |
| Approach: | They propose a benchmark to evaluate instruction-following in mathematical reasoning tasks. |
| Outcome: | The proposed model degrades in instruction adherence when generation length increases, but can partially recover obedience, despite increasing generation length. |
There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods emphasize selecting one golden knowledge given a particular dialogue context, overlooking the one-to-many phenomenon in dialogue. |
| Approach: | They propose to use a multi-reference dataset to assess the one-to-many efficacy of existing KGC models. |
| Outcome: | The proposed model improves the mapping relationship between multiple knowledge and multiple responses by optimizing the model in a wake-sleep style. |
SORTIE: Dependency-Aware Symbolic Reasoning for Logical Data-to-text Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies on logical data-to-text generation rely on neural language models to generate the final table description, but they have difficulty working out key entities in the description. |
| Approach: | They propose a symbolic reasoning framework that reasons out each entity in the table description with a table-compatible programming language. |
| Outcome: | The proposed framework outperforms existing methods on three datasets and three backbones with an absolute improvement of 5.7%11.5% on SP-Acc. |
Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction (2024.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained language models may not follow human instructions and produce toxic, hallucinated, or biased content. |
| Approach: | They propose a disperse-then-merge framework that dispersers instruction-following data into portions and trains multiple sub-models using different data portions. |
| Outcome: | The proposed framework outperforms data curation and training regularization on standard knowledge and reasoning benchmarks. |
The Best of Both Worlds: Combining Parallel and Sequential Inference Scaling via Aggregation Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Empirical results show that AFT-trained models achieve substantial gains with test-time scaling. |
| Approach: | They introduce a supervised fine-tuning paradigm where models synthesize multiple draft responses into a single, refined answer. |
| Outcome: | Empirical results show that AFT-trained models outperform baseline models while eliminating external guidance. |
On the Compositional Generalization in Versatile Open-domain Dialogue (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to multi-task learning suffer from interference among datasets or fail to effectively reuse knowledge and skills learned from other datasets. |
| Approach: | They propose a sparsely activated modular network with a well-rounded set of operators and instantiate each operator with an independent module. |
| Outcome: | The proposed model outperforms state-of-the-art supervised approaches on 4 datasets with only 10% training data thanks to the modular architecture and multi-task learning. |