Qi Jia

Siyu Ren, Qi Jia, Kenny Q. Zhu

The quadratic complexity of the attention module makes it gradually become the bulk of compute in Transformer-based LLMs during generation. Moreover, the excessive key-value cache that arises when dealing with long inputs also brings severe issues on memory footprint and inference latency. In this work, we propose a plug-and-play approach that is able to incrementally compress the intermediate activation of a specified span of tokens into compact ones, thereby reducing both memory and computational cost when processing subsequent context. Experiments on both in-domain language modeling and zero-shot open-ended document generation demonstrate the advantage of our approach over sparse attention baselines in terms of fluency, n-gram matching, and semantic similarity. At last, we comprehensively profile the benefit of context compression on improving the system throughout. Code is available at https://github.com/DRSY/KV_Compression.

Qi Jia, Siyu Ren, Yizhu Liu et al.

Despite tremendous improvements in natural language generation, summarization models still suffer from the unfaithfulness issue. Previous work evaluates faithfulness either using models trained on the other tasks or in-domain synthetic data, or prompting a large model such as ChatGPT. This paper proposes to do zero-shot faithfulness evaluation simply with a moderately-sized foundation language model. We introduce a new metric FFLM, which is a combination of probability changes based on the intuition that prefixing a piece of text that is consistent with the output will increase the probability of predicting the output. Experiments show that FFLM performs competitively with or even outperforms ChatGPT on both inconsistency detection and faithfulness rating with 24x fewer parameters. FFLM also achieves improvements over other strong baselines.

Qi Jia, Yizhu Liu, Siyu Ren et al.

Multi-turn response selection is a task designed for developing dialogue agents. The performance on this task has a remarkable improvement with pre-trained language models. However, these models simply concatenate the turns in dialogue history as the input and largely ignore the dependencies between the turns. In this paper, we propose a dialogue extraction algorithm to transform a dialogue history into threads based on their dependency relations. Each thread can be regarded as a self-contained sub-dialogue. We also propose Thread-Encoder model to encode threads and candidates into compact representations by pre-trained Transformers and finally get the matching score through an attention layer. The experiments show that dependency relations are helpful for dialogue context understanding, and our model outperforms the state-of-the-art baselines on both DSTC7 and DSTC8*, with competitive results on UbuntuV2.

Qi Jia, Mengxue Zhang, Shengyao Zhang et al.

Matching question-answer relations between two turns in conversations is not only the first step in analyzing dialogue structures, but also valuable for training dialogue systems. This paper presents a QA matching model considering both distance information and dialogue history by two simultaneous attention mechanisms called mutual attention. Given scores computed by the trained model between each non-question turn with its candidate questions, a greedy matching strategy is used for final predictions. Because existing dialogue datasets such as the Ubuntu dataset are not suitable for the QA matching task, we further create a dataset with 1,000 labeled dialogues and demonstrate that our proposed model outperforms the state-of-the-art and other strong baselines, particularly for matching long-distance QA pairs.