NeurIPS 2024 accepted papers on Large language models
612 of the 4,560 papers accepted at NeurIPS 2024 (13.4%) are on Large language models, compared with 6.0% at NeurIPS 2023. They fall into 7 topics.
See NeurIPS 2024 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Language, safety and interpretability · 202 papers
- LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential Recommendation (61 citations)
- Tree of Attacks: Jailbreaking Black-Box LLMs Automatically (55 citations)
- Membership Inference Attacks against Fine-tuned Large Language Models via Self-prompt Calibration (28 citations)
- CultureLLM: Incorporating Cultural Differences into Large Language Models (26 citations)
- WildGuard: Open One-stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs (25 citations)
- The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale (22 citations)
- Many-shot Jailbreaking (21 citations)
- HaloScope: Harnessing Unlabeled LLM Generations for Hallucination Detection (19 citations)
- A StrongREJECT for Empty Jailbreaks (18 citations)
- Cross-Care: Assessing the Healthcare Implications of Pre-training Data on Language Model Bias (18 citations)
- On the Worst Prompt Performance of Large Language Models (16 citations)
- RouterDC: Query-Based Router by Dual Contrastive Learning for Assembling Large Language Models (16 citations)
and 190 more in the interactive map.
Agents, code and math · 131 papers
- SGLang: Efficient Execution of Structured Language Model Programs (109 citations)
- Gorilla: Large Language Model Connected with Massive APIs (93 citations)
- MDAgents: An Adaptive Collaboration of LLMs for Medical Decision-Making (58 citations)
- ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution (53 citations)
- CTIBench: A Benchmark for Evaluating LLMs in Cyber Threat Intelligence (40 citations)
- Unveiling Causal Reasoning in Large Language Models: Reality or Mirage? (24 citations)
- CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs (18 citations)
- SocraticLM: Exploring Socratic Personalized Teaching with Large Language Models (18 citations)
- Evaluating the World Model Implicit in a Generative Model (14 citations)
- IaC-Eval: A Code Generation Benchmark for Cloud Infrastructure-as-Code Programs (14 citations)
- Embodied Agent Interface: Benchmarking LLMs for Embodied Decision Making (13 citations)
- SemCoder: Training Code Language Models with Comprehensive Semantics Reasoning (13 citations)
and 119 more in the interactive map.
Instruction and fine-tuning · 77 papers
- Refusal in Language Models Is Mediated by a Single Direction (57 citations)
- HydraLoRA: An Asymmetric LoRA Architecture for Efficient Fine-Tuning (24 citations)
- Recursive Introspection: Teaching Language Model Agents How to Self-Improve (15 citations)
- Mercury: A Code Efficiency Benchmark for Code Large Language Models (13 citations)
- Large Scale Transfer Learning for Tabular Data via Language Modeling (12 citations)
- BLoB: Bayesian Low-Rank Adaptation by Backpropagation for Large Language Models (11 citations)
- ReFT: Representation Finetuning for Language Models (10 citations)
- Safe LoRA: The Silver Lining of Reducing Safety Risks when Finetuning Large Language Models (8 citations)
- Customizing Language Models with Instance-wise LoRA for Sequential Recommendation (7 citations)
- SmallToLarge (S2L): Scalable Data Selection for Fine-tuning Large Language Models by Summarizing Training Trajectories of Small Models (7 citations)
- Alignment for Honesty (6 citations)
- BackdoorAlign: Mitigating Fine-tuning based Jailbreak Attack with Backdoor Enhanced Safety Alignment (6 citations)
and 65 more in the interactive map.
Retrieval and knowledge · 71 papers
- HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (105 citations)
- LLM-Check: Investigating Detection of Hallucinations in Large Language Models (41 citations)
- Make Your LLM Fully Utilize the Context (40 citations)
- Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs (38 citations)
- Large Language Models as Urban Residents: An LLM Agent Framework for Personal Mobility Generation (36 citations)
- RankRAG: Unifying Context Ranking with Retrieval-Augmented Generation in LLMs (30 citations)
- Chain of Agents: Large Language Models Collaborating on Long-Context Tasks (27 citations)
- FinBen: A Holistic Financial Benchmark for Large Language Models (27 citations)
- KnowGPT: Knowledge Graph based Prompting for Large Language Models (23 citations)
- CRAG - Comprehensive RAG Benchmark (21 citations)
- RAGChecker: A Fine-grained Framework for Diagnosing Retrieval-Augmented Generation (20 citations)
- Large Language Models-guided Dynamic Adaptation for Temporal Knowledge Graph Reasoning (17 citations)
and 59 more in the interactive map.
Alignment and preferences · 56 papers
- LLM Evaluators Recognize and Favor Their Own Generations (65 citations)
- SimPO: Simple Preference Optimization with a Reference-Free Reward (39 citations)
- Large Language Model Unlearning (34 citations)
- Mobility-LLM: Learning Visiting Intentions and Travel Preference from Human Mobility Data with Large Language Models (30 citations)
- On Softmax Direct Preference Optimization for Recommendation (23 citations)
- Questioning the Survey Responses of Large Language Models (22 citations)
- UltraMedical: Building Specialized Generalists in Biomedicine (16 citations)
- The PRISM Alignment Dataset: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models (15 citations)
- Aligning LLM Agents by Learning Latent Preference from User Edits (13 citations)
- $\beta$-DPO: Direct Preference Optimization with Dynamic $\beta$ (11 citations)
- Iterative Reasoning Preference Optimization (9 citations)
- Rule Based Rewards for Language Model Safety (8 citations)
and 44 more in the interactive map.
Reasoning and chain of thought · 41 papers
- MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark (82 citations)
- SnapKV: LLM Knows What You are Looking for Before Generation (51 citations)
- Chain-of-Thought Reasoning Without Prompting (36 citations)
- ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search (21 citations)
- Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs (19 citations)
- Efficient LLM Scheduling by Learning to Rank (19 citations)
- Instance-adaptive Zero-shot Chain-of-Thought Prompting (13 citations)
- AutoMix: Automatically Mixing Language Models (12 citations)
- Chain of Thoughtlessness? An Analysis of CoT in Planning (9 citations)
- SELF-DISCOVER: Large Language Models Self-Compose Reasoning Structures (8 citations)
- Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaboration (7 citations)
- Unlocking the Capabilities of Thought: A Reasoning Boundary Framework to Quantify and Optimize Chain-of-Thought (7 citations)
and 29 more in the interactive map.
AI and society · 34 papers
- BetterBench: Assessing AI Benchmarks, Uncovering Issues, and Establishing Best Practices (21 citations)
- RedPajama: an Open Dataset for Training Large Language Models (19 citations)
- Croissant: A Metadata Format for ML-Ready Datasets (18 citations)
- Consent in Crisis: The Rapid Decline of the AI Data Commons (13 citations)
- CulturePark: Boosting Cross-cultural Understanding in Large Language Models (12 citations)
- Are More LLM Calls All You Need? Towards the Scaling Properties of Compound AI Systems (8 citations)
- Navigating the Maze of Explainable AI: A Systematic Approach to Evaluating Methods and Metrics (8 citations)
- Evaluating Large Vision-and-Language Models on Children's Mathematical Olympiads (6 citations)
- Utilizing Human Behavior Modeling to Manipulate Explanations in AI-Assisted Decision Making: The Good, the Bad, and the Scary (6 citations)
- Indoor Air Quality Dataset with Activities of Daily Living in Low to Middle-income Communities (5 citations)
- Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs (5 citations)
- OlympicArena: Benchmarking Multi-discipline Cognitive Reasoning for Superintelligent AI (5 citations)
and 22 more in the interactive map.