TiFA @ ICML 2024

14 papers

Top Research Directions

The 24 research directions with TiFA @ ICML 2024 papers, ranked by paper count. Share is the fraction of the edition’s papers in the direction.

#DirectionPapersShare
1Activation Steering for Inference Time Model Control17.1%
2Direct Preference Optimization17.1%
3Game Theoretic Formulations for Adversarial Robustness17.1%
4Learning from Incomplete Multimodal Data Streams17.1%
5Loss Landscape Flattening for Sequential Learning17.1%
6Low Rank Adaptation for Efficient Model Customization17.1%
7Overcoming the Digital to Physical Domain Gap17.1%
8Parameter-Efficient Multimodal Adaptation17.1%
9Preserving Safety Alignment During Downstream Fine-Tuning17.1%
10Sample Efficient Learning for Specialized Domains17.1%
11Scaling Multimodal Fusion Beyond Pairwise Interactions17.1%
12Solving Complex Games via Zero-Sum Reductions17.1%
13Sparse Temporal Adversarial Perturbations17.1%
14Synchronized Multimodal Adversarial Attacks17.1%
15Task Activation via In Context Demonstrations17.1%
16Task Order Sensitivity in Continual Learning17.1%
17The Causal Disconnect in Language Model Reasoning17.1%
18The Fragility of Imperceptible Image Protections17.1%
19The Inverse Scaling of Reasoning Faithfulness17.1%
20The Sufficiency of Memoryless Strategies17.1%
21The Vulnerability of Static Defenses to Adaptive Adversaries17.1%
22Visual Grounding and Hallucination in Vision Language Models17.1%
23Why Adversarial Attacks Transfer Across Language Models17.1%
24Why Text to Image Models Amplify Societal Stereotypes17.1%

Download all directions as CSV

Institutions

Institutions credited by TiFA @ ICML 2024 papers: a paper credits every institution named in its affiliations, once. 29 institutions share 14 credited papers.

#InstitutionPapersShare
1Harvard University 214.3%
2McGill University 214.3%
3Meta (United States) 214.3%
4Mila - Quebec Artificial Intelligence Institute 214.3%
5Tencent (China) 214.3%
6Tsinghua University 214.3%
701zna6t55 17.1%
803n4cgv79 17.1%
9Korea Advanced Institute of Science and Technology 17.1%
10Michigan State University 17.1%
11Nanyang Technological University 17.1%
12NEC (United States) 17.1%
13New York University 17.1%
14Peng Cheng Laboratory 17.1%
15Queen Mary University of London 17.1%
16Rutgers, The State University of New Jersey 17.1%
17Shanghai Artificial Intelligence Laboratory 17.1%
18Shanghai Jiao Tong University 17.1%
19Sun Yat-sen University 17.1%
20University College London 17.1%
21University College London Hospitals NHS Foundation Trust 17.1%
22University of California, Berkeley 17.1%
23University of California, Santa Barbara 17.1%
24University of Cambridge 17.1%
25University of Illinois Urbana-Champaign 17.1%
26University of Massachusetts Amherst 17.1%
27University of Massachusetts System 17.1%
28University of Oxford 17.1%
29Université de Montréal 17.1%

Download all institutions as CSV

About this data

Papers. TiFA @ ICML 2024 lists 14 papers on Lacuna, duplicate versions counted once; rejected and withdrawn submissions are left out.

Directions. Directions are Lacuna’s clustering of the concept elements extracted from each paper, so a paper can belong to several and direction counts add up to more than the paper count. A direction’s share is its papers divided by the edition’s papers.

Institutions. A paper credits each institution named in its affiliations, resolved to a ROR record, once. 14 of 14 papers (100.0%) credit at least one institution and are the base for institution shares; the rest have no resolved affiliation yet.

Data. Counts come from Lacuna’s data build of 2026-11-18 and change as papers are added. Tables as CSV: directions · institutions. Cite as: Lacuna, “TiFA @ ICML 2024”, https://lacuna.tiptreesystems.com/venue/tifa-icml/2024.

Top Authors

AuthorPapers
Adina Williams2
Yang Bai2
Abdul W Basit1
Abhishek Sureddy1
Adam Daneshmend1
Adam Ibrahim1
Adriana Romero-Soriano1
Alessandro Abate1
Alexis Roger1
Alicia Yi Sun1
Andreas Demosthenous1
Aviral Kumar1
Boryeong Cho1
Brando Miranda1
Buck Shlegeris1
Charlie Griffin1
Chirag Agarwal1
Claire Tomlin1
Daniel Z Kaplan1
Dan Ley1

Impactful Papers

Papers from TiFA @ ICML 2024 that shaped research clusters.

Influential

Papers

Title
Bias Begets Bias: The Impact of Biased Embeddings on Diffusion Models
Borrowing Treasures from Neighbors: In-Context Learning for Multimodal Learning with Missing Modalities and Data Scarcity
Chained Tuning Leads to Biased Forgetting
Decomposed evaluations of geographic disparities in text-to-image models
Games for AI-Control: Models of Safety Evaluations of AI Deployment Protocols
On the Difficulty of Faithful Chain-of-Thought Reasoning in Large Language Models
On the Multi-modal Vulnerability of Diffusion Models
Physical Backdoor Attack can Jeopardize Driving with Vision-Large-Language Models
Towards Adversarially Robust Vision-Language Models: Insights from Design Choices and Prompt Formatting Techniques
TrustAgent: Towards Safe and Trustworthy LLM-based Agents through Agent Constitution
Unfamiliar Finetuning Examples Control How Language Models Hallucinate
VACoDe: Visual Augmented Contrastive Decoding
Video Watermarking: Safeguarding Your Video from (Unauthorized) Annotations by Video-based LLMs
Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?