Papers › TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy

TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy

25 Jan 2024IEEE/ACM Transactions on Audio, Speech, and Language Processing 2024 1arXiv:2402.01714archive 2025-07-28

Vibhav Agarwal, Sourav Ghosh, Harichandana BSS, Himanshu Arora, Barath Raj Kandur Raja

Data-to-text (D2T) generation is a crucial task in many natural language understanding (NLU) applications and forms the foundation of task-oriented dialog systems. In the context of conversational AI solutions that can work directly with local data on the user's device, architectures utilizing large pre-trained language models (PLMs) are impractical for on-device deployment due to a high memory footprint. To this end, we propose TrICy, a novel lightweight framework for an enhanced D2T task that generates text sequences based on the intent in context and may further be guided by user-provided triggers. We leverage an attention-copy mechanism to predict out-of-vocabulary (OOV) words accurately. Performance analyses on E2E NLG dataset (BLEU: 66.43%, ROUGE-L: 70.14%), WebNLG dataset (BLEU: Seen 64.08%, Unseen 52.35%), and our Custom dataset related to text messaging applications, showcase our architecture's effectiveness. Moreover, we show that by leveraging an optional trigger input, data-to-text generation quality increases significantly and achieves the new SOTA score of 69.29% BLEU for E2E NLG. Furthermore, our analyses show that TrICy achieves at least 24% and 3% improvement in BLEU and METEOR respectively over LLMs like GPT-3, ChatGPT, and Llama 2. We also demonstrate that in some scenarios, performance improvement due to triggers is observed even when they are absent in training.

Paper PDF Conference PDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Data-to-Text Generation Natural Language Understanding Text Generation

Results from the paper archive 2025-07-28

Task	Dataset	Model	Metric	Value	Rank at snapshot	Leaderboard	Report
Data-to-Text Generation	E2E NLG Challenge	TrICy (trK = 0)	BLEU	66.43	#3 of 11	Archive leaderboard	report
Data-to-Text Generation	E2E NLG Challenge	TrICy (trK = 0)	Number of parameters (M)	4.7	#3 of 11	Archive leaderboard	report
Data-to-Text Generation	E2E NLG Challenge	TrICy (trK = 0)	ROUGE-L	70.14	#3 of 11	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = trk* = 0.24)	BLEU	64.73	#10 of 20	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = trk* = 0.24)	METEOR	45.53	#10 of 20	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = trk* = 0.24)	Number of parameters (M)	6.2	#10 of 20	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = 0)	BLEU	64.08	#12 of 20	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = 0)	METEOR	45.23	#12 of 20	Archive leaderboard	report
Data-to-Text Generation	WebNLG	TrICy (trK = 0)	Number of parameters (M)	6.2	#12 of 20	Archive leaderboard	report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Adam Attention Attention Dropout BPE Cosine Annealing Dense Connections Dropout GPT-3 Layer Normalization Linear Layer Linear Warmup With Cosine Annealing Multi-Head Attention Residual Connection Softmax Weight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections