Reading Group: Natural Language Processing

This is the page of the internal reading group of the NLP Group. It is currently invitation-only. In the reading group, recent research papers from the field of natural language processing (NLP) are discussed, with varying themes.

• Winter 2024/25. Selected state-of-the art NLP techniques in depth

• Summer 2024. Selected papers from recent NLP research

• Winter 2023/24. State-of-the-art techniques in NLP

• Summer 2023. NLP papers outside the NLP community

• Winter 2022/23. Best papers from major NLP conferences

• Summer 2022. Selected papers from our main research areas

• Winter 2021/22. Self-assessment of recent papers by the NLP Group

• Summer 2021. Selected papers from our main research areas

• Winter 2020/21. Self-assessment of recent papers by the NLP Group

• Rimsky et al. (2024). Steering Llama 2 via Contrastive Activation Addition, https://aclanthology.org/2024.acl-long.828/ (December 10, 2024; technique: activation-based steering)
   
• Turner et al. (2023). Steering Language Models with Activation Engineering, https://arxiv.org/pdf/2308.10248 (November 26, 2024; technique: activation-based steering)
   
• Rafailov et al. (2023). Direct Preference Optimization: Your Language Model is Secretly a Reward Model, https://proceedings.neurips.cc/paper_files/paper/2023/file/a85b405ed65c6477a4fe8302b5e06ce7-Paper-Conference.pdf (November 12, 2024; technique: reinforcement learning)
   
• Zheng et al. (2023). Secrets of RLHF in Large Language Models Part I: PPO, https://arxiv.org/abs/2307.04964 (October 29, 2024; technique: reinforcement learning)
   
• Li et al. (2016). Deep Reinforcement Learning for Dialogue Generation, https://aclanthology.org/D16-1127/ (October 15, 2024; technique: reinforcement learning)

• Zhang et al. (2024). Fair Abstractive Summarization of Diverse Perspectives, https://aclanthology.org/2024.naacl-long.187/ (July 16, 2024)
   
• Zhou et al. (2024). Diffusion-NAT: Self-Prompting Discrete Diffusion for Non-Autoregressive Text Generation, https://aclanthology.org/2024.eacl-long.86 (June 18, 2024)
   
• Konen et al. (2024). Style Vectors for Steering Generative Large Language Models, https://aclanthology.org/2024.findings-eacl.52 (June 4, 2024)
   
• Kim et al. (2024). Prometheus: Inducing Fine-grained Evaluation Capability in Language Models, https://arxiv.org/abs/2310.08491 (May 7, 2024)
   
• Qiao et al. (2023). Reasoning with Language Model Prompting: A Survey, https://aclanthology.org/2023.acl-long.294/ (April 23, 2024)
   
• Fillippova (2020). Controlled Hallucinations:Learning to Generate Faithfully from Noisy Data, https://aclanthology.org/2020.findings-emnlp.76/ (April 9, 2024)
   

• Lewis et al. (2020). Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks, https://proceedings.neurips.cc/paper/2020/file/6b493230205f780e1bc26945df7481e5-Paper.pdf (February 6, 2024)
   
• Hewitt et al. (2023). Backpack Language Models, https://aclanthology.org/2023.acl-long.506.pdf (January 23, 2024)
   
• Li et al. (2022). Diffusion-LM Improves Controllable Text Generation,  https://proceedings.neurips.cc/paper_files/paper/2022/hash/1be5bc25d50895ee656b8c2d9eb89d6a-Abstract-Conference.html (January 9, 2024)
   
• Wei et al. (2022). Chain-of-Thought Prompting Elicits Reasoning in Large Language Models, https://proceedings.neurips.cc/paper_files/paper/2022/hash/9d5609613524ecf4f15af0f7b31abca4-Abstract-Conference.html  (November 28, 2023)
   
• Pfeiffer et al. (2021). AdapterFusion: Non-Destructive Task Composition for Transfer Learning, https://aclanthology.org/2021.eacl-main.39.pdf (November 14, 2023)
   
• Wang et al. (2023). Self-Instruct: Aligning Language Models with Self-Generated Instructions, https://aclanthology.org/2023.acl-long.754.pdf (October 17, 2023)
   

• Islam et al. (2023). PATRON: Perspective-Aware Multitask Model for Referring Expression Grounding Using Embodied Multimodal Cues. https://ojs.aaai.org/index.php/AAAI/article/view/25177 (July 13, 2023, area: multimodal)
   
• Li et al. (2022). SimStu-Transformer: A Transformer-Based Approach to Simulating Student Behaviour. https://link.springer.com/chapter/10.1007/978-3-031-11647-6_67 (June 30, 2023, area: education)
   
• Niu et al. (2022). AttExplainer: Explain Transformer via Attention by Reinforcement Learning. https://www.ijcai.org/proceedings/2022/0102  (June 16, 2023, area: AI)
   
• Nakashima et al. (2020). Virus database annotations assist in tracing information on patients infected with emerging pathogens. https://www.sciencedirect.com/science/article/pii/S2352914820306067 (June 2, 2023, area: medicine)
   
• Jakesh et al. (2023). Co-Writing with Opinionated Language Models Affects Users’ Views, https://dl.acm.org/doi/10.1145/3544548.3581196 (May 11, 2023, area: HCI)
   
• Agarwal et al. (2022). GraphNLI: A Graph-based Natural Language Inference Model for Polarity Prediction in Online Debates, https://dl.acm.org/doi/pdf/10.1145/3485447.3512144 (April 27, 2023, area: web)
   
• Lin et al. (2022). What Makes the Story Forward? Inferring Commonsense Explanations as Prompts for Future Event Generation, https://dl.acm.org/doi/10.1145/3477495.3532080 (April 13, 2023, area: information retrieval)

• Wen et al. (2015). Semantically Conditioned LSTM-based Natural Language Generation for Spoken Dialogue Systems, https://aclanthology.org/D15-1199 (February 28, 2023)

• Anonymized. Unpublished paper (February 14, 2023)

• Andreas et al. (2016). Learning to Compose Neural Networks for Question Answering, https://aclanthology.org/N16-1181 (January 31, 2023)

• Kottur et al. (2017). Natural Language Does Not Emerge ‘Naturally’ in Multi-Agent Dialog, https://aclanthology.org/D17-1321, (January 17, 2023)

• Peters et al. (2018). Deep contextualized word representations, https://aclanthology.org/N18-1202 (January 3, 2023)

• Moon et al. (2019). OpenDialKG: Explainable Conversational Reasoning with Attention-based Walks over Knowledge Graphs, https://aclanthology.org/P19-1081 (December 20, 2022)

• Bender and Koller (2020). Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data, https://aclanthology.org/2020.acl-main.463 (December 6, 2022)

• Qin et al. (2021). Learning How to Ask: Querying LMs with Mixtures of Soft Prompts, https://aclanthology.org/2021.naacl-main.410 (November 22, 2022)

• Lu et al. (2022). NEUROLOGIC A*esque Decoding: Constrained Text Generation with Lookahead Heuristics,  https://aclanthology.org/2022.naacl-main.57 (November 8, 2022)

• Bianci and Hovy (2022). On the Gap between Adoption and Understanding in NLP

• Ke et al. (2022). CTRLEval: An Unsupervised Reference-Free Metric for Evaluating Controlled Text Generation

• Barzilay and Lapata (2005). Modeling Local Coherence: An Entity-based Approach

• Ye et al. (2021). CROSSFIT: A Few-shot Learning Challenge for Cross-task Generalization in NLP

• Jin et al. (2022). Logical Fallacy Detection

• Sun et al. (2021). Do Long-Range Language Models Actually Use Long-Range Context?

• Hanawa et al. (2021). Exploring Methods for Generating Feedback Comments for Writing Learning 

• Schick and Schütze (2021). Generating Datasets with Pretrained Language Models

• Paul and Frank (2020). Social Commonsense Reasoning with Multi-Head Knowledge Attention

• Barrow et al. (2021). Syntopical Graphs for Computational Argumentation Tasks

• Chen et al. (2021). Controlled Neural Sentence-Level Reframing of News Articles

• Alshomary et al. (2021). Argument Undermining: Counter-Argument Generation by Attacking Weak Premises

• Wachsmuth et al. (2014). Modeling Review Argumentation for Robust Sentiment Analysis

• Spliethöver and Wachsmuth (2021). Bias Silhouette Analysis: Towards Assessing the Quality of Bias Metrics for Word Embedding Models

• Gurcke et al. (2021). Assessing the Sufficiency of Arguments through Conclusion Generation

• Prabhumoye et al. (2021). Case Study: Deontological Ethics in NLP

• Chen et al. (2018). Application of Sentiment Analysis to Language Learning

• Blodgett et al. (2020). Language (Technology) is Power: A Critical Survey of “Bias” in NLP

• Orbach and Goldberg (2020). Facts2Story: Controlling Text Generation by Key Facts

• Lertvittayakumjorn et al. (2020). FIND: Human-in-the-Loop Debugging Deep Text Classifiers

• Hegel et al. (2020). Substance over Style: Document-Level Targeted Content Transfer

• Tan et al. (2020). Summarizing Text on Any Aspects: A Knowledge-Informed Weakly-Supervised Approach

• Spliethöver and Wachsmuth (2020). Argument from Old Man’s View: Assessing Social Bias in Argumentation

• Alshomary et al. (2020). Belief-based Generation of Argumentative Claims

• Chen et al. (2020). Detecting Media Bias in News Articles using Gaussian Bias Distributions

• Wachsmuth and Werner (2020). Intrinsic Quality Assessment of Arguments

• Stein et al. (2014). Generating Acrostics via Paraphrasing and Heuristic Search

• Wachsmuth (2015). Towards Ad-hoc Large-Scale Text Mining, Chapter 1 (Introduction)

• Alshomary et al. (2019). Wikipedia Text Reuse: Within and Without

• Spliethöver et al. (2019). Is It Worth the Attention? A Comparative Evaluation of Attention Layers for Argument Unit Segmentation

• Wachsmuth et al. (2018). Retrieval of the Best Counterargument without Prior Topic Knowledge

• Chen et al. (2018). Learning to Flip the Bias of News Headlines