Research map / Graph learning / Knowledge and temporal graphs
Graph-based dialog systems
Use graph structures for dialog reasoning and language model context.
94 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Knowledge and temporal graphs, keyword cluster “scene · image · visual”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
27 papers in 2023–24 and 16 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 10.5% to 5.7%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Inferring Interpersonal Relations in Narrative Summaries (AAAI 2016 · 34 citations)
- 2017 · Coupled Multi-Layer Attentions for Co-Extraction of Aspect and Opinion Terms (AAAI 2017 · 448 citations)
- 2018 · A Neural Transition-Based Approach for Semantic Dependency Graph Parsing (AAAI 2018 · 63 citations)
- 2019 · Neighbourhood Watch: Referring Expression Comprehension via Language-Guided Graph Attention Networks (CVPR 2019 · 292 citations)
- 2020 · Schema-Guided Multi-Domain Dialogue State Tracking with Graph Attention Neural Networks (AAAI 2020 · 145 citations)
- 2021 · Document-Level Relation Extraction with Reconstruction (AAAI 2021 · 107 citations)
- 2022 · Knowledge Bridging for Empathetic Dialogue Generation (AAAI 2022 · 112 citations)
- 2023 · Multivariate, Multi-Frequency and Multimodal: Rethinking Graph Neural Networks for Emotion Recognition in Conversation (CVPR 2023 · 83 citations)
- 2024 · G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering (NeurIPS 2024 · 74 citations)
- 2025 · Revisiting Multimodal Emotion Recognition in Conversation from the Perspective of Graph Spectrum (AAAI 2025 · 19 citations)
- 2026 · RoSE: A Role Correlation Structure-Enhanced Model for Multi-Event Argument Extraction (AAAI 2026 · 1 citation)
Earliest papers here
- Inferring Interpersonal Relations in Narrative Summaries (AAAI 2016 · 34 citations)
- Extracting Biomolecular Interactions Using Semantic Parsing of Biomedical Text (AAAI 2016 · 22 citations)
- Dependency Tree Representations of Predicate-Argument Structures (AAAI 2016 · 5 citations)
What has been tried
83 papers here report 904 measured changes (a component added, removed or swapped, compared with a baseline): helped 78%, no clear effect 4%, hurt 14%, mixed 4%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add graph encoder | WebQSP · Hit@1 | 54.62 ± 0.78 -> 70.49 ± 1.21 | Helped | G-Retriever: Retrieval-Augmented Generation for Textual Graph Understanding and Question Answering (NeurIPS 2024 · 74 citations) |
| Add positional embedding | SciERC · Strict Relation F1 (REL+) | 36.4 -> 38.6 | Helped | An Autoregressive Text-to-Graph Framework for Joint Entity and Relation Extraction (AAAI 2024 · 42 citations) |
| Add kNN inference to FGM | Multimodal sarcasm detection test set · Accuracy | 89.01 -> 89.86 | Helped | G^2SAM: Graph-Based Global Semantic Awareness Method for Multimodal Sarcasm Detection (AAAI 2024 · 22 citations) |
| add speaker embedding | IEMOCAP and MELD · W-Acc and W-F1 | IEMOCAP: 70.3 (↓2.8) W-Acc, 70.6 (↓2.7) W-F1; MELD: 65.4 (↓2.7) W-Acc, 64.6 (↓4.4) W-F1 | Helped | Revisiting Multimodal Emotion Recognition in Conversation from the Perspective of Graph Spectrum (AAAI 2025 · 19 citations) |
| add graph matching score computation | HM3D IIN · SR | 54.9 -> 60.2 | Helped | UniGoal: Towards Universal Zero-shot Goal-oriented Navigation (CVPR 2025 · 15 citations) |
| add the R-S graph | DiaASQ Chinese · quadruple F1 | 37.02 -> 40.34 | Helped | Harnessing Holistic Discourse Features and Triadic Interaction for Sentiment Quadruple Extraction in Dialogues (AAAI 2024 · 14 citations) |
| Replace graph question encoder with application question encoder | GQA Edge Existence · accuracy | 42.8 -> 60.8 | Helped | Talk like a Graph: Encoding Graphs for Large Language Models (ICLR 2024 · 13 citations) |
| replace attention-based claim-evidence interaction with mean pooling | FEVEROUS dev · Feverous score | 0.3613 vs 0.3714 | Hurt | Heterogeneous Graph Reasoning for Fact Checking over Texts and Tables (AAAI 2024 · 12 citations) |
| use the goal interaction graph instead of removing graph learning | DuRecDial · goal entity F1 | 66.75 -> 82.75 | Helped | A Goal Interaction Graph Planning Framework for Conversational Recommendation (AAAI 2024 · 9 citations) |
| add the heterogeneous-relation-aware graph module | ECE-CCKS test · ECE F1 | 53.10 -> 54.34 | Helped | A Joint Framework with Heterogeneous-Relation-Aware Graph and Multi-Channel Label Enhancing Strategy for Event Causality Extraction (AAAI 2024 · 6 citations) |
| add token-level context sentences for token embeddings | Genia · F1 | 82.97 -> 83.36 | Helped | Span Graph Transformer for Document-Level Named Entity Recognition (AAAI 2024 · 6 citations) |
| Add the Heterogeneous Graph Construction module | EgoTV novel steps · F1-score | 50.7 -> 58.2 | Helped | PHGC: Procedural Heterogeneous Graph Completion for Natural Language Task Verification in Egocentric Videos (CVPR 2025 · 5 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 43 papers; from OpenAlex and the papers' affiliations).
- Microsoft · 2 papers
- Beijing Institute of Technology · 2 papers
- JD AI Research · 2 papers
- Chinese Academy of Sciences · 2 papers
- University of Chinese Academy of Sciences · 2 papers
- Harbin Institute of Technology · 2 papers
- Wuhan University · 2 papers
- Tianjin University · 2 papers
- National University of Singapore · 2 papers
- University of Electronic Science and Technology of China · 2 papers
Use it
Download BibTeX of 39 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Graph reasoning for visual recognition (96 papers)
- Heterogeneous graph learning (66 papers)
- Social and behavior graphs (87 papers)
- Scene graph generation (70 papers)
Elsewhere in the topic
- Continuous-time dynamic graph learning (132 papers)
- Knowledge graph embeddings (110 papers)
- Temporal knowledge graph reasoning (84 papers)
- Temporal graph neural architectures (69 papers)
- Logical query answering and rule reasoning (68 papers)
- Spatio-temporal graph traffic forecasting (60 papers)
- Inductive knowledge graph completion (38 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).