Total relevant papers: 6

Paper selection prompt and criteria at the bottom

Table of contents with paper titles:

0. Measuring Non-Adversarial Reproduction of Training Data in Large Language Models Authors: Michael Aerni, Javier Rando, Edoardo Debenedetti, Nicholas Carlini, Daphne Ippolito, Florian Tram`er

1. Mitigating Hallucination in Multimodal Large Language Model via Hallucination-targeted Direct Preference Optimization Authors: Yuhan Fu, Ruobing Xie, Xingwu Sun, Zhanhui Kang, Xirong Li

2. Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization Authors: Weiyun Wang, Zhe Chen, Wenhai Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Jinguo Zhu, Xizhou Zhu, Lewei Lu, Yu Qiao, Jifeng Dai

3. Large Language Models as User-Agents for Evaluating Task-Oriented-Dialogue Systems Authors: Taaha Kazi, Ruiliang Lyu, Sizhe Zhou, Dilek Hakkani-Tur, Gokhan Tur

4. Entropy and type-token ratio in gigaword corpora Authors: Pablo Rosillo-Rodes, Maxi San Miguel, David Sanchez

5. Layer Importance and Hallucination Analysis in Large Language Models via Enhanced Activation Variance-Sparsity Authors: Zichen Song, Sitan Huang, Yuxin Wu, Zhongfeng Kang

---

ArXiv ID: 2411.10242 Authors: Michael Aerni, Javier Rando, Edoardo Debenedetti, Nicholas Carlini, Daphne Ippolito, Florian Tram`er

Abstract: arXiv:2411.10242v1 Announce Type: new Abstract: Large language models memorize parts of their training data. Memorizing short snippets and facts is required to answer questions about the world and to be fluent in any language. But models have also been shown to reproduce long verbatim sequences of memorized text when prompted by a motivated adversary. In this work, we investigate an intermediate regime of memorization that we call non-adversarial reproduction, where we quantify the overlap between model responses and pretraining data when responding to natural and benign prompts. For a variety of innocuous prompt categories (e.g., writing a letter or a tutorial), we show that up to 15% of the text output by popular conversational language models overlaps with snippets from the Internet. In worst cases, we find generations where 100% of the content can be found exactly online. For the same tasks, we find that human-written text has far less overlap with Internet data. We further study whether prompting strategies can close this reproduction gap between models and humans. While appropriate prompting can reduce non-adversarial reproduction on average, we find that mitigating worst-case reproduction of training data requires stronger defenses -- even for benign interactions.

Comment: Relevant to criterion 2 as it discusses non-adversarial reproduction of training data in language models, which is related to test set contamination. Relevance: 8 Novelty: 6

---

ArXiv ID: 2411.10436 Authors: Yuhan Fu, Ruobing Xie, Xingwu Sun, Zhanhui Kang, Xirong Li

Abstract: arXiv:2411.10436v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) are known to hallucinate, which limits their practical applications. Recent works have attempted to apply Direct Preference Optimization (DPO) to enhance the performance of MLLMs, but have shown inconsistent improvements in mitigating hallucinations. To address this issue more effectively, we introduce Hallucination-targeted Direct Preference Optimization (HDPO) to reduce hallucinations in MLLMs. Unlike previous approaches, our method tackles hallucinations from their diverse forms and causes. Specifically, we develop three types of preference pair data targeting the following causes of MLLM hallucinations: (1) insufficient visual capabilities, (2) long context generation, and (3) multimodal conflicts. Experimental results demonstrate that our method achieves superior performance across multiple hallucination evaluation datasets, surpassing most state-of-the-art (SOTA) methods and highlighting the potential of our approach. Ablation studies and in-depth analyses further confirm the effectiveness of our method and suggest the potential for further improvements through scaling up.

Comment: This paper does not match any of the specified criteria closely. Relevance: 3 Novelty: 5

---

ArXiv ID: 2411.10442 Authors: Weiyun Wang, Zhe Chen, Wenhai Wang, Yue Cao, Yangzhou Liu, Zhangwei Gao, Jinguo Zhu, Xizhou Zhu, Lewei Lu, Yu Qiao, Jifeng Dai

Abstract: arXiv:2411.10442v1 Announce Type: new Abstract: Existing open-source multimodal large language models (MLLMs) generally follow a training process involving pre-training and supervised fine-tuning. However, these models suffer from distribution shifts, which limit their multimodal reasoning, particularly in the Chain-of-Thought (CoT) performance. To address this, we introduce a preference optimization (PO) process to enhance the multimodal reasoning capabilities of MLLMs. Specifically, (1) on the data side, we design an automated preference data construction pipeline to create MMPR, a high-quality, large-scale multimodal reasoning preference dataset. and (2) on the model side, we explore integrating PO with MLLMs, developing a simple yet effective method, termed Mixed Preference Optimization (MPO), which boosts multimodal CoT performance. Our approach demonstrates improved performance across multiple benchmarks, particularly in multimodal reasoning tasks. Notably, our model, InternVL2-8B-MPO, achieves an accuracy of 67.0 on MathVista, outperforming InternVL2-8B by 8.7 points and achieving performance comparable to the 10x larger InternVL2-76B. We hope this study could inspire further advancements in MLLMs. Code, data, and model shall be publicly released.

Comment: This paper does not match any of the specified criteria closely. Relevance: 3 Novelty: 5

---

ArXiv ID: 2411.09972 Authors: Taaha Kazi, Ruiliang Lyu, Sizhe Zhou, Dilek Hakkani-Tur, Gokhan Tur

Abstract: arXiv:2411.09972v1 Announce Type: new Abstract: Traditionally, offline datasets have been used to evaluate task-oriented dialogue (TOD) models. These datasets lack context awareness, making them suboptimal benchmarks for conversational systems. In contrast, user-agents, which are context-aware, can simulate the variability and unpredictability of human conversations, making them better alternatives as evaluators. Prior research has utilized large language models (LLMs) to develop user-agents. Our work builds upon this by using LLMs to create user-agents for the evaluation of TOD systems. This involves prompting an LLM, using in-context examples as guidance, and tracking the user-goal state. Our evaluation of diversity and task completion metrics for the user-agents shows improved performance with the use of better prompts. Additionally, we propose methodologies for the automatic evaluation of TOD models within this dynamic framework.

Comment: This paper does not match any of the specified criteria closely. Relevance: 3 Novelty: 4

---

ArXiv ID: 2411.10227 Authors: Pablo Rosillo-Rodes, Maxi San Miguel, David Sanchez

Abstract: arXiv:2411.10227v1 Announce Type: new Abstract: Lexical diversity measures the vocabulary variation in texts. While its utility is evident for analyses in language change and applied linguistics, it is not yet clear how to operationalize this concept in a unique way. We here investigate entropy and text-token ratio, two widely employed metrics for lexical diversities, in six massive linguistic datasets in English, Spanish, and Turkish, consisting of books, news articles, and tweets. These gigaword corpora correspond to languages with distinct morphological features and differ in registers and genres, thus constituting a diverse testbed for a quantitative approach to lexical diversity. Strikingly, we find a functional relation between entropy and text-token ratio that holds across the corpora under consideration. Further, in the limit of large vocabularies we find an analytical expression that sheds light on the origin of this relation and its connection with both Zipf and Heaps laws. Our results then contribute to the theoretical understanding of text structure and offer practical implications for fields like natural language processing.

Comment: This paper does not match any of the specified criteria closely. Relevance: 3 Novelty: 4

---

ArXiv ID: 2411.10069 Authors: Zichen Song, Sitan Huang, Yuxin Wu, Zhongfeng Kang

Abstract: arXiv:2411.10069v1 Announce Type: new Abstract: Evaluating the importance of different layers in large language models (LLMs) is crucial for optimizing model performance and interpretability. This paper first explores layer importance using the Activation Variance-Sparsity Score (AVSS), which combines normalized activation variance and sparsity to quantify each layer's contribution to overall model performance. By ranking layers based on AVSS and pruning the least impactful 25%, our experiments on tasks such as question answering, language modeling, and sentiment classification show that over 90% of the original performance is retained, highlighting potential redundancies in LLM architectures. Building on AVSS, we propose an enhanced version tailored to assess hallucination propensity across layers (EAVSS). This improved approach introduces Hallucination-Specific Activation Variance (HSAV) and Hallucination-Specific Sparsity (HSS) metrics, allowing precise identification of hallucination-prone layers. By incorporating contrastive learning on these layers, we effectively mitigate hallucination generation, contributing to more robust and efficient LLMs(The maximum performance improvement is 12%). Our results on the NQ, SciQ, TriviaQA, TruthfulQA, and WikiQA datasets demonstrate the efficacy of this method, offering a comprehensive framework for both layer importance evaluation and hallucination mitigation in LLMs.

Comment: This paper does not match any of the specified criteria closely. Relevance: 3 Novelty: 4

---

---

1. New methodological improvements to RLHF or instruction-following which are specific fine-tuning steps that are taken to make language models better at following user instructions across a range of tasks.
   • Relevant: papers that discuss specific methods like RLHF, or instruction-tuning datasets, improving these methods, or analyzing them. Usually these papers will explicitly mention RLHF, instruction-following or instruction-tuning.
   • Not relevant: papers about adaptation to some task. Simply following instructions or inputs are not sufficient.
2. Shows new powerful test set contamination or membership inference methods for language models. Test set contamination is the phenomenon where a language model observes a benchmark dataset during pretraining.
   • Relevant: test statistics that can detect contamination of benchmarks in language models. statistics that can provide guarantees are more interesting. membership inference methods that are general enough to apply to language models are also relevant.
   • Not relevant: any papers that do not consider language models, or that do not consider test set contamination.
3. Shows a significant advance in the performance of diffusion language models.
   • Relevant: papers that study language models that are also diffusion models. Continuous diffusions are even more relevant, while discrete diffusions are less so.
   • Not relevant: papers about image diffusions like DALL-E or Stable Diffusion, or papers that do not explicitly mention language models or applications to text.
4. Describes new paradigms to evaluating open-ended text generation. Evaluating the outputs of language models is hard, especially in open-ended settings like for chatbots.
   • Relevant: papers that fundamentally rethink language model evaluation -- especially by accounting for subjectivity or using adversaries.
   • Not relevant: specific evaluations for specific tasks, identifying new properties or flaws of language models, or simply collecting new data.
5. Conducts surveys or provides data into real-world usage and safety properties of language models.
   • Relevant: papers that create new datasets or surveys on real-world usage of language models.
   • Not relevant: papers that apply language models to new real-world tasks.
6. Studies 'scaling laws' in the context of neural networks. Scaling laws refer to the very clear power-law relationship between the size or computational power used to train a model and the performance of that model.
   • Relevant: theoretical or conceptual explanation behind scaling laws for language models.
   • Not relevant: papers that have experiments at different model scales (but do not explicitly fit a scaling law) or papers that mention scaling laws, but the scaling laws are not the central subject of the paper

In suggesting papers to your friend, remember that he enjoys papers on statistical machine learning, and generative modeling in natural language processing. Your friend also likes learning about surprising empirical results in language models, as well as clever statistical tricks. He does not want to read papers that are about primarily applications of methods to specific domains.