DongqiFu_UIUC Profile Banner
Dongqi Fu Profile
Dongqi Fu

@DongqiFu_UIUC

Followers
1K
Following
433
Media
7
Statuses
55

Research Scientist at @AIatMeta | Ph.D. from @IllinoisCDS | Working on #GeometricDeepLearning #SequenceModeling #ProbabilisticGraphicalModel

Joined January 2017
Don't wanna be here? Send us removal request.
@HEI
Natural Language Processing Papers
2 months
Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation.
arxiv.org
Modern long-context large language models (LLMs) perform well on synthetic "needle-in-a-haystack" (NIAH) benchmarks, but such tests overlook how noisy contexts arise from biased retrieval and...
0
1
2
@Jiaru_Zou
Jiaru "Rubin" Zou
2 months
๐ŸšจReleasing ๐—ฅ๐—”๐—š ๐—ผ๐˜ƒ๐—ฒ๐—ฟ ๐—ง๐—ฎ๐—ฏ๐—น๐—ฒ๐˜€: ๐—›๐—ถ๐—ฒ๐—ฟ๐—ฎ๐—ฟ๐—ฐ๐—ต๐—ถ๐—ฐ๐—ฎ๐—น ๐— ๐—ฒ๐—บ๐—ผ๐—ฟ๐˜† ๐—œ๐—ป๐—ฑ๐—ฒ๐˜…, ๐— ๐˜‚๐—น๐˜๐—ถ-๐—ฆ๐˜๐—ฎ๐—ด๐—ฒ ๐—ฅ๐—ฒ๐˜๐—ฟ๐—ถ๐—ฒ๐˜ƒ๐—ฎ๐—น, ๐—ฎ๐—ป๐—ฑ ๐—•๐—ฒ๐—ป๐—ฐ๐—ต๐—บ๐—ฎ๐—ฟ๐—ธ๐—ถ๐—ป๐—ด. ๐Ÿš€ We push RAG to Multi-tables! ๐ŸŒCode: https://t.co/kppOT6uXBC ๐Ÿ“„Paper: https://t.co/KclevgAf6z
1
4
8
@DongqiFu_UIUC
Dongqi Fu
2 months
RAG over Tables ๐Ÿ›ซ
@Jiaru_Zou
Jiaru "Rubin" Zou
2 months
๐ŸšจReleasing ๐—ฅ๐—”๐—š ๐—ผ๐˜ƒ๐—ฒ๐—ฟ ๐—ง๐—ฎ๐—ฏ๐—น๐—ฒ๐˜€: ๐—›๐—ถ๐—ฒ๐—ฟ๐—ฎ๐—ฟ๐—ฐ๐—ต๐—ถ๐—ฐ๐—ฎ๐—น ๐— ๐—ฒ๐—บ๐—ผ๐—ฟ๐˜† ๐—œ๐—ป๐—ฑ๐—ฒ๐˜…, ๐— ๐˜‚๐—น๐˜๐—ถ-๐—ฆ๐˜๐—ฎ๐—ด๐—ฒ ๐—ฅ๐—ฒ๐˜๐—ฟ๐—ถ๐—ฒ๐˜ƒ๐—ฎ๐—น, ๐—ฎ๐—ป๐—ฑ ๐—•๐—ฒ๐—ป๐—ฐ๐—ต๐—บ๐—ฎ๐—ฟ๐—ธ๐—ถ๐—ป๐—ด. ๐Ÿš€ We push RAG to Multi-tables! ๐ŸŒCode: https://t.co/kppOT6uXBC ๐Ÿ“„Paper: https://t.co/KclevgAf6z
0
0
3
@_reachsumit
Sumit
2 months
Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation @Mufei_Li et al. present a benchmark built on Wikipedia's hyperlink network. ๐Ÿ“ https://t.co/l0fpCNAbx6 ๐Ÿ‘จ๐Ÿฝโ€๐Ÿ’ป https://t.co/iJZWY1gQcC
github.com
Haystack Engineering: Context Engineering for Heterogeneous and Agentic Long-Context Evaluation - Graph-COM/HaystackCraft
1
4
10
@DongqiFu_UIUC
Dongqi Fu
4 months
Very inspiring work! Congrats!
@_Violet24K_
Zihao Li
4 months
๐ŸŒ Flow Matching Meets Biology and Life Science: A Survey Flow matching is emerging as a powerful generative paradigm. We comprehensively review its foundations and applications across biology & life science๐Ÿงฌ ๐Ÿ“šPaper: https://t.co/ynsegKOgXz ๐Ÿ’ปResource:
0
0
0
@kthrn_uiuc
Katherine Tieu
5 months
โ€ผ๏ธWe propose to learn temporal positional encodings for spatial-temporal graphs, L-STEP, in our #ICML2025 paper. Even simple MLPs can achieve leading performance in various temporal link prediction settings! ๐Ÿ“„ Paper:ย  https://t.co/ioapcELCD5 ๐Ÿ’ปย Code:ย  https://t.co/sQWMlBZ1N6
0
2
11
@DongqiFu_UIUC
Dongqi Fu
6 months
Interesting to see the analysis between Inference scaling and LLM safety!
@GaotangLi
Gaotang Li
6 months
๐Ÿ˜ฒ Not only reasoning?! Inference scaling can now boost LLM safety! ๐Ÿš€ Introducing Saffron-1: - Reduces attack success rate from 66% to 17.5% - Uses only 59.7 TFLOP compute - Counters latest jailbreak attacks - No model finetuning On the AI2 Refusals benchmark. ๐Ÿ“– Paper:
0
0
0
@DongqiFu_UIUC
Dongqi Fu
7 months
Thrilled to have 3 #ICML and 1 #ACL accepted! Congrats all collaborators ๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰ Studying graph representation learning, multimodal alignment, and generative material design๐Ÿ˜„ Stay tuned๐Ÿค”
0
3
11
@DongqiFu_UIUC
Dongqi Fu
8 months
Identify Invariant Subgraphs for Spatial-Temporal OOD Problems ๐Ÿ‘๐Ÿ‘๐Ÿ‘
@kthrn_uiuc
Katherine Tieu
8 months
Solving Out-of-Distribution problem of Spatial-Temporal Graphsโ“ Check out our #AISTATS2025 paper, our method distinguishes invariant components during training and makes link predictions robust to distribution shifts Paper: https://t.co/6ZPxK4TZHI
0
0
10
@ideaisailuiuc
iDEA-iSAIL Group@UIUC
8 months
We'll present 4 papers and 1 keynote talk at #ICLR2025. Prof. Jingrui He and Prof. Hanghang Tong will be at the conference. Let's connect! โ˜•๏ธ
0
3
8
@DongqiFu_UIUC
Dongqi Fu
9 months
๐Ÿ”ฅvarious graph types and various evaluation metrics
@ideaisailuiuc
iDEA-iSAIL Group@UIUC
9 months
๐Ÿ”ฌGraph Self-Supervised Learning Toolkit ๐Ÿ”ฅWe release PyG-SSL, offering a unified framework of 10+ self-supervised choices to pretrain your graph foundation models. ๐Ÿ“œPaper: https://t.co/fwlWTsmquK ๐Ÿ’ปCode: https://t.co/oNaz18zCht Have fun!
0
0
3
@DongqiFu_UIUC
Dongqi Fu
10 months
Beyond numerical-only time series models, letโ€™s see how language can help time series forecasting and imputation๐Ÿคน
@_Violet24K_
Zihao Li
10 months
๐Ÿ“ˆ Your time-series-paired texts are secretly a time series! ๐Ÿ™Œ Real-world time series (stock prices) and texts (financial reports) share similar periodicity and spectrum, unlocking seamless multimodal learning using existing TS models. ๐Ÿ”ฌ Read more: https://t.co/YYAH4J1Wne
0
0
1
@DongqiFu_UIUC
Dongqi Fu
10 months
๐Ÿ’ก How can we describe a graph to LLMs ? ๐Ÿง For example, G(n, p) uses number of nodes and connection probability to describe a graph. ๐Ÿ“‘ Please check out our survey, What Do LLMs Need to Understand Graphs: A Survey of Parametric Representation of Graphs https://t.co/6bNIAwH5nG
0
6
12
@DongqiFu_UIUC
Dongqi Fu
10 months
Excited to know our #ICLR2025 paper is selected as Spotlight ๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰ We studied how to generate large-scale temporal heterogeneous graphs balancing privacy, utility, and efficiency We also found diffusion model may not be the solution to answer them all https://t.co/yiAaG71pvp
openreview.net
Nowadays, temporal heterogeneous graphs attract much research and industrial attention for building the next-generation Relational Deep Learning models and applications, due to their informative...
0
2
13
@tempgraph_rg
temporal graph learning reading group
10 months
This thursday, Feb 13th, 11am ET, at the Reading group, @kthrn_uiuc will present: Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed (NeurIPS 2024). Looking forward to seeing you there! ๐ŸŽ‰๐ŸŽ‰ Zoom link on website https://t.co/oKBW6mUhNj
openreview.net
_Graph Neural Tangent Kernel_ (GNTK) fuses graph neural networks and graph kernels, simplifies the process of graph representation learning, interprets the training dynamics of graph neural...
@kthrn_uiuc
Katherine Tieu
1 year
Can we obtain temporal graph neural representation without neural networkโ“ Check out our #NeurIPS2024 paper, Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed ๐ŸŽ‰ Paper: https://t.co/3u0bobtWJ9 Code: https://t.co/JYrCtiUjxm
0
3
8
@DongqiFu_UIUC
Dongqi Fu
11 months
2 papers accepted by #ICLR2025, 1 paper by #AISTATS2025, and 1 paper by #KDD2025, congrats all collaborators ๐ŸŽ‰ TL;DR 1๏ธโƒฃ How to generate temporal heterogeneous graphs 2๏ธโƒฃ How to tokenize a graph 3๏ธโƒฃ How to infer invariant subgraphs 4๏ธโƒฃ How to compress an evolving knowledge graph
0
3
13
@DongqiFu_UIUC
Dongqi Fu
1 year
๐Ÿง When deepening is unavoidable, how to avoid oversmoothing of deeper GNNs? ๐Ÿ’ก Topology-guided contrastive learning can help ๐Ÿ“ข Check our new #TMLR paper, DrGNN: Deep Residual Graph Neural Network with Contrastive Learning Link:
0
0
1
@DongqiFu_UIUC
Dongqi Fu
1 year
๐Ÿง How deep a GNN should be? ๐Ÿ’ก Number of layers is related the diameter of the input graph ๐Ÿง When deepening is unavoidable, how to avoid dimensional collapse of deeper GNNs? ๐Ÿ’ก Topology-guided residual connection of neural layers can help
1
3
8
@kthrn_uiuc
Katherine Tieu
1 year
Can we obtain temporal graph neural representation without neural networkโ“ Check out our #NeurIPS2024 paper, Temporal Graph Neural Tangent Kernel with Graphon-Guaranteed ๐ŸŽ‰ Paper: https://t.co/3u0bobtWJ9 Code: https://t.co/JYrCtiUjxm
1
2
10
@ideaisailuiuc
iDEA-iSAIL Group@UIUC
1 year
We will present 6 papers at #NeurIPS2024 covering active learning, graph learning, time series and trustworthy ML. Come and have a chat during the poster sessions!๐Ÿš€
0
4
6