Towards Verifiable Text Generation with Evolving Memory and Self-Reflection

Large Language Models (LLMs) face several challenges, including the tendency to produce incorrect outputs, known as hallucination. An effective solution is verifiable text generation, which prompts LLMs to generate content with citations for accuracy verification. However, verifiable text generation is non-trivial due to the focus-shifting phenomenon, the dilemma between the precision and scope in document retrieval, and the intricate reasoning required to discern the relationship between the claim and citations. In this paper, we present VTG, an innovative approach for Verifiable Text Generation with evolving memory and self-reflection. VTG maintains evolving long short-term memory to retain both valuable documents and up-to-date documents. Active retrieval and diverse query generation are utilized to enhance both the precision and scope of the retrieved documents. Furthermore, VTG features a two-tier verifier and an evidence finder, enabling rethinking and reflection on the relationship between the claim and citations. We conduct extensive experiments on five datasets across three knowledge-intensive tasks and the results reveal that VTG significantly outperforms existing baselines.

Unveiling the Implicit Toxicity in Large Language Models

The open-endedness of large language models (LLMs) combined with their impressive capabilities may lead to new safety issues when being exploited for malicious use. While recent studies primarily focus on probing toxic outputs that can be easily detected with existing toxicity classifiers, we show that LLMs can generate diverse implicit toxic outputs that are exceptionally difficult to detect via simply zero-shot prompting. Moreover, we propose a reinforcement learning (RL) based attacking method to further induce the implicit toxicity in LLMs. Specifically, we optimize the language model with a reward that prefers implicit toxic outputs to explicit toxic and non-toxic ones. Experiments on five widely-adopted toxicity classifiers demonstrate that the attack success rate can be significantly improved through RL fine-tuning. For instance, the RL-finetuned LLaMA-13B model achieves an attack success rate of 90.04% on BAD and 62.85% on Davinci003. Our findings suggest that LLMs pose a significant threat in generating undetectable implicit toxic outputs. We further show that fine-tuning toxicity classifiers on the annotated examples from our attacking method can effectively enhance their ability to detect LLM-generated implicit toxic language. The code is publicly available at https://github.com/thu-coai/Implicit-Toxicity.

Robust Domain Misinformation Detection via Multi-modal Feature Alignment

Social media misinformation harms individuals and societies and is potentialized by fast-growing multi-modal content (i.e., texts and images), which accounts for higher "credibility" than text-only news pieces. Although existing supervised misinformation detection methods have obtained acceptable performances in key setups, they may require large amounts of labeled data from various events, which can be time-consuming and tedious. In turn, directly training a model by leveraging a publicly available dataset may fail to generalize due to domain shifts between the training data (a.k.a. source domains) and the data from target domains. Most prior work on domain shift focuses on a single modality (e.g., text modality) and ignores the scenario where sufficient unlabeled target domain data may not be readily available in an early stage. The lack of data often happens due to the dynamic propagation trend (i.e., the number of posts related to fake news increases slowly before catching the public attention). We propose a novel robust domain and cross-modal approach (\textbf{RDCM}) for multi-modal misinformation detection. It reduces the domain shift by aligning the joint distribution of textual and visual modalities through an inter-domain alignment module and bridges the semantic gap between both modalities through a cross-modality alignment module. We also propose a framework that simultaneously considers application scenarios of domain generalization (in which the target domain data is unavailable) and domain adaptation (in which unlabeled target domain data is available). Evaluation results on two public multi-modal misinformation detection datasets (Pheme and Twitter Datasets) evince the superiority of the proposed model. The formal implementation of this paper can be found in this link: https://github.com/less-and-less-bugs/RDCM

When is Off-Policy Evaluation Useful? A Data-Centric Perspective

Evaluating the value of a hypothetical target policy with only a logged dataset is important but challenging. On the one hand, it brings opportunities for safe policy improvement under high-stakes scenarios like clinical guidelines. On the other hand, such opportunities raise a need for precise off-policy evaluation (OPE). While previous work on OPE focused on improving the algorithm in value estimation, in this work, we emphasize the importance of the offline dataset, hence putting forward a data-centric framework for evaluating OPE problems. We propose DataCOPE, a data-centric framework for evaluating OPE, that answers the questions of whether and to what extent we can evaluate a target policy given a dataset. DataCOPE (1) forecasts the overall performance of OPE algorithms without access to the environment, which is especially useful before real-world deployment where evaluating OPE is impossible; (2) identifies the sub-group in the dataset where OPE can be inaccurate; (3) permits evaluations of datasets or data-collection strategies for OPE problems. Our empirical analysis of DataCOPE in the logged contextual bandit settings using healthcare datasets confirms its ability to evaluate both machine-learning and human expert policies like clinical guidelines.

AI-accelerated Discovery of Altermagnetic Materials

Altermagnetism, a new magnetic phase, has been theoretically proposed and experimentally verified to be distinct from ferromagnetism and antiferromagnetism. Although altermagnets have been found to possess many exotic physical properties, the very limited availability of known altermagnetic materials (e.g., 14 confirmed materials) hinders the study of such properties. Hence, discovering more types of altermagnetic materials is crucial for a comprehensive understanding of altermagnetism and thus facilitating new applications in the next-generation information technologies, e.g., storage devices and high-sensitivity sensors. Here, we report 25 new altermagnetic materials that cover metals, semiconductors, and insulators, discovered by an AI search engine unifying symmetry analysis, graph neural network pre-training, optimal transport theory, and first-principles electronic structure calculation. The wide range of electronic structural characteristics reveals that various novel physical properties manifest in these newly discovered altermagnetic materials, e.g., anomalous Hall effect, anomalous Kerr effect, and topological property. Noteworthy, we discovered 8 i-wave altermagnetic materials for the first time. Overall, the AI search engine performs much better than human experts and suggests a set of new altermagnetic materials with unique properties, outlining its potential for accelerated discovery of the materials with targeting properties.

Character-level Chinese Backpack Language Models

The Backpack is a Transformer alternative shown to improve interpretability in English language modeling by decomposing predictions into a weighted sum of token sense components. However, Backpacks' reliance on token-defined meaning raises questions as to their potential for languages other than English, a language for which subword tokenization provides a reasonable approximation for lexical items. In this work, we train, evaluate, interpret, and control Backpack language models in character-tokenized Chinese, in which words are often composed of many characters. We find that our (134M parameter) Chinese Backpack language model performs comparably to a (104M parameter) Transformer, and learns rich character-level meanings that log-additively compose to form word meanings. In SimLex-style lexical semantic evaluations, simple averages of Backpack character senses outperform input embeddings from a Transformer. We find that complex multi-character meanings are often formed by using the same per-character sense weights consistently across context. Exploring interpretability-through control, we show that we can localize a source of gender bias in our Backpacks to specific character senses and intervene to reduce the bias.

Accountability in Offline Reinforcement Learning: Explaining Decisions with a Corpus of Examples

Learning transparent, interpretable controllers with offline data in decision-making systems is an essential area of research due to its potential to reduce the risk of applications in real-world systems. However, in responsibility-sensitive settings such as healthcare, decision accountability is of paramount importance, yet has not been adequately addressed by the literature. This paper introduces the Accountable Offline Controller (AOC) that employs the offline dataset as the Decision Corpus and performs accountable control based on a tailored selection of examples, referred to as the Corpus Subset. ABC operates effectively in low-data scenarios, can be extended to the strictly offline imitation setting, and displays qualities of both conservation and adaptability. We assess ABC's performance in both simulated and real-world healthcare scenarios, emphasizing its capability to manage offline control tasks with high levels of performance while maintaining accountability. Keywords: Interpretable Reinforcement Learning, Explainable Reinforcement Learning, Reinforcement Learning Transparency, Offline Reinforcement Learning, Batched Control.

Reinforcement Learning in the Era of LLMs: What is Essential? What is needed? An RL Perspective on RLHF, Prompting, and Beyond

Recent advancements in Large Language Models (LLMs) have garnered wide attention and led to successful products such as ChatGPT and GPT-4. Their proficiency in adhering to instructions and delivering harmless, helpful, and honest (3H) responses can largely be attributed to the technique of Reinforcement Learning from Human Feedback (RLHF). In this paper, we aim to link the research in conventional RL to RL techniques used in LLM research. Demystify this technique by discussing why, when, and how RL excels. Furthermore, we explore potential future avenues that could either benefit from or contribute to RLHF research. Highlighted Takeaways: 1. RLHF is Online Inverse RL with Offline Demonstration Data. 2. RLHF $>$ SFT because Imitation Learning (and Inverse RL) $>$ Behavior Cloning (BC) by alleviating the problem of compounding error. 3. The RM step in RLHF generates a proxy of the expensive human feedback, such an insight can be generalized to other LLM tasks such as prompting evaluation and optimization where feedback is also expensive. 4. The policy learning in RLHF is more challenging than conventional problems studied in IRL due to their high action dimensionality and feedback sparsity. 5. The main superiority of PPO over off-policy value-based methods is its stability gained from (almost) on-policy data and conservative policy updates.

Query-Dependent Prompt Evaluation and Optimization with Offline Inverse RL

In this study, we aim to enhance the arithmetic reasoning ability of Large Language Models (LLMs) through zero-shot prompt optimization. We identify a previously overlooked objective of query dependency in such optimization and elucidate two ensuing challenges that impede the successful and economical design of prompt optimization techniques. One primary issue is the absence of an effective method to evaluate prompts during inference when the golden answer is unavailable. Concurrently, learning via interactions with the LLMs to navigate the expansive natural language prompting space proves to be resource-intensive. To address this, we introduce Prompt-OIRL, which harnesses offline inverse reinforcement learning to draw insights from offline prompting demonstration data. Such data exists as by-products when diverse prompts are benchmarked on open-accessible datasets. With Prompt-OIRL, the query-dependent prompt optimization objective is achieved by first learning an offline reward model. This model can evaluate any query-prompt pairs without accessing LLMs. Subsequently, a best-of-N strategy is deployed to recommend the optimal prompt. Our experimental evaluations across various LLM scales and arithmetic reasoning datasets underscore both the efficacy and economic viability of the proposed approach.