Conversational Recommender Systems (CRS) typically require domain-specific dialogue data, which is costly, scarce, and often unavailable in new domains. We conduct a systematic empirical study of zero-data CRS bootstrapping: generating synthetic conversational supervision from non-conversational signals—item reviews, metadata, and user-item interactions—without any in-domain dialogue corpus. We compare two information-theoretic selection strategies, Jensen-Shannon diversity and Fisher information, across domain signals, model architectures, datasets, and fine-tuning paradigms. Our results show that domain-grounded synthetic data consistently outperforms zero-shot prompting and naive synthetic baselines; active selection improves data efficiency over random sampling; metadata and collaborative filtering signals each improve selection quality; and, in low-resource settings, synthetic data can outperform scarce real dialogues while further complementing them. These findings establish non-conversational domain signals as a viable path toward building CRS without conversational training data. The code is available at https://anonymous.4open.science/r/zero_data_crs/ .

Large language models improve final-answer accuracy through extended chain-of-thought reasoning, but often spend tokens inefficiently and offer little inference-time control. Existing efficient reasoning methods control thinking length by shortening, early-stopping, or compressing traces, leaving how the model thinks implicit. In this paper, we propose Agentic Chain-of-Thought Steering (ACTS), which formulates reasoning steering as a Markov decision process where a controller agent adaptively steers a frozen reasoner during inference. At each step, the controller observes the reasoning trace and remaining thinking budget, then issues a steering action consisting of a reasoning strategy and a steering phrase that initiates the next reasoner step. This enables budget-aware strategy control for efficient reasoning while preserving the reasoner’s generation continuity. We initialize the controller agent from our constructed synthetic steering trajectories with multi-budget augmentation, and further optimize it via reinforcement learning with budget-conditioned reward shaping. Experiments across multiple benchmarks show that ACTS achieves competitive accuracy with substantial token savings, and enables controllable accuracy–token trade-offs across different reasoners and tasks. The code is available at https://github.com/Andree-9/ACTS.

The advent of autonomous agents is transforming interactions with Graphical User Interfaces (GUIs) by employing natural language as a powerful intermediary. Despite the predominance of supervised fine-tuning (SFT) methods in current GUI agents for achieving spatial localization, these methods face substantial challenges due to their limited capacity to accurately perceive positional data. Existing strategies, such as reinforcement learning, often fail to assess positional accuracy effectively, thereby restricting their utility. In response, we introduce Location Preference Optimization (LPO), a novel approach that leverages locational data to optimize interaction preferences. LPO uses information entropy to predict interaction positions by focusing on zones rich in information. Besides, we further introduce a dynamic location reward function based on physical distance, reflecting the varying importance of interaction positions. Supported by Group Relative Preference Optimization (GRPO), LPO facilitates an extensive exploration of GUI environments and significantly enhances interaction precision. Comprehensive experiments demonstrate LPO’s superior performance, achieving SOTA results across both offline benchmarks and real-world online evaluations.

Small Language Models (SLMs) have become increasingly important due to their efficiency and performance to perform various language tasks with minimal computational resources, making them ideal for various settings including on-device, mobile, edge devices, among many others. In this article, we present a comprehensive survey on SLMs, focusing on their architectures, training techniques, and model compression techniques. We propose a novel taxonomy for categorizing the methods used to optimize SLMs, including model compression, pruning, and quantization techniques. We summarize the benchmark datasets that are useful for benchmarking SLMs along with the evaluation metrics commonly used. Additionally, we highlight key open challenges that remain to be addressed. Our survey aims to serve as a valuable resource for researchers and practitioners interested in developing and deploying small yet efficient language models.

Retrieval-augmented large language models (LLMs) have been remarkably competent in various NLP tasks. However, it was observed by previous works that retrieval is not always helpful, especially when the LLM is already knowledgable on the query to answer. Motivated by this, Adaptive Retrieval-Augmented Generation (ARAG) studies retrieving only when the knowledge asked by the query is absent in the LLM. Previous works of ARAG either require accessing the pre-training corpus or prompting with additional model inferences. Aiming to avoid such drawbacks, we propose to determine whether the model is knowledgeable on a query via inspecting the (contextualized) pre-trained token embeddings of LLMs. We hypothesize that such embeddings capture rich information on the model’s intrinsic knowledge base, which enables an efficient way of judging the necessity to retrieve from an external corpus. Extensive experiments demonstrate our ARAG approach’s superior performance across various benchmarks.

Chain-of-Thought (CoT) has been a widely adopted prompting method, eliciting impressive reasoning abilities of Large Language Models (LLMs). Inspired by the sequential thought structure of CoT, a number of Chain-of-X (CoX) methods have been developed to address challenges across diverse domains and tasks. In this paper, we provide a comprehensive survey of Chain-of-X methods for LLMs in different contexts. Specifically, we categorize them by taxonomies of nodes, i.e., the X in CoX, and application tasks. We also discuss the findings and implications of existing CoX methods, as well as potential future directions. Our survey aims to serve as a detailed and up-to-date resource for researchers seeking to apply the idea of CoT to broader scenarios.

Existing LLM-based agents have achieved strong performance on held-in tasks, but their generalizability to unseen tasks remains poor. Hence, some recent work focus on fine-tuning the policy model with more diverse tasks to improve the generalizability. In this work, we find that finetuning a reward model to guide the policy model is more robust than directly finetuning the policy model.Based on this finding, we propose AgentRM, a 8B generalizable reward model, to guide the policy model for effective test-time search.We comprehensively investigate three approaches to construct the reward model, including explicit reward modeling, implicit reward modeling and LLM-as-a-judge.We then use AgentRM to guide the answer generation with Best-of-N sampling and beam search.We show that AgentRM is robust to paraphrasings of task instructions and can generalize to unseen tasks that require novel optimal behavior.Through extensive evaluation across nine tasks spanning four categories, AgentRM enhances the non-finetuned 8B policy model by 8.8 points on average, surpassing the top general agent by 4.0.Moreover, it demonstrates weak-to-strong generalization, yielding greater improvement on more powerful policy models.As for the specializability, AgentRM can also boost a finetuned policy model and outperform the top specialized agent by 11.4 on three held-in tasks.Further analysis verifies its effectiveness in test-time scaling.We release the code and data at https://github.com/thunlp/AgentRM.

Large Language Models (LLMs) have shown propensity to generate hallucinated outputs, i.e., texts that are factually incorrect or unsupported. Existing methods for alleviating hallucinations typically require costly human annotations to identify and correct hallucinations in LLM outputs. Moreover, most of these methods focus on a specific type of hallucination, e.g., entity or token errors, which limits their effectiveness in addressing various types of hallucinations exhibited in LLM outputs. To our best knowledge, in this paper we propose the first active learning framework to alleviate LLM hallucinations, reducing costly human annotations of hallucination needed. By measuring fine-grained hallucinations from errors in semantic frame, discourse and content verifiability in text summarization, we propose HAllucination Diversity-Aware Sampling (HADAS) to select diverse hallucinations for annotations in active learning for LLM finetuning. Extensive experiments on three datasets and different backbone models demonstrate advantages of our method in effectively and efficiently mitigating LLM hallucinations.

Large language models (LLMs) often generate biased outputs containing offensive, toxic, or stereotypical text. Existing LLM alignment methods such as reinforcement learning from human feedback (RLHF) alleviate biases primarily based on reward signals from current model outputs without considering the source of biases. In this work, to explore how biases are formed, we revisit LLMs’ text generation from a causal perspective. We identify pretraining data and input prompts, which contain semantic correlations of textual phrases, as two confounders between LLMs and model outputs causing biases. Inspired by our causal view, we leverage the reward model in RL alignment as an instrumental variable to perform causal intervention on LLMs. Utilizing the reward difference between an initial LLM and intervened LLM as interventional feedback to guide RL finetuning, we propose Causality-Aware Alignment (CAA) for LLM debiasing. Experiments on two text generation tasks with three different alignment objectives demonstrate the advantages of our method in aligning LLMs to generate less biased and safer outputs.

Text style transfer (TST) without parallel data has achieved some practical success. However, most of the existing unsupervised text style transfer methods suffer from (i) requiring massive amounts of non-parallel data to guide transferring different text styles. (ii) colossal performance degradation when fine-tuning the model in new domains. In this work, we propose DAML-ATM (Domain Adaptive Meta-Learning with Adversarial Transfer Model), which consists of two parts: DAML and ATM. DAML is a domain adaptive meta-learning approach to learn general knowledge in multiple heterogeneous source domains, capable of adapting to new unseen domains with a small amount of data. Moreover, we propose a new unsupervised TST approach Adversarial Transfer Model (ATM), composed of a sequence-to-sequence pre-trained language model and uses adversarial style training for better content preservation and style transfer. Results on multi-domain datasets demonstrate that our approach generalizes well on unseen low-resource domains, achieving state-of-the-art results against ten strong baselines.

Traditional methods for named entity recognition (NER) classify mentions into a fixed set of pre-defined entity types. However, in many real-world scenarios, new entity types are incrementally involved. To investigate this problem, continual learning is introduced for NER. However, the existing method depends on the relevance between tasks and is prone to inter-type confusion. In this paper, we propose a novel two-stage framework Learn-and-Review (L&R) for continual NER under the type-incremental setting to alleviate the above issues. Specifically, for the learning stage, we distill the old knowledge from teacher to a student on the current dataset. For the reviewing stage, we first generate synthetic samples of old types to augment the dataset. Then, we further distill new knowledge from the above student and old knowledge from the teacher to get an enhanced student on the augmented dataset. This stage has the following advantages: (1) The synthetic samples mitigate the gap between the old and new task and thus enhance the further distillation; (2) Different types of entities are jointly seen during training which alleviates the inter-type confusion. Experimental results show that L&R outperforms the state-of-the-art method on CoNLL-03 and OntoNotes-5.0.

Multi-hop Question Answering is an agent task for testing the reasoning ability. With the development of pre-trained models, the implicit reasoning ability has been surprisingly improved and can even surpass human performance. However, the nature of the black box hinders the construction of explainable intelligent systems. Several researchers have explored explainable neural-symbolic reasoning methods based on question decomposition techniques. The undifferentiable symbolic operations and the error propagation in the reasoning process lead to poor performance. To alleviate it, we propose a simple yet effective Global Differentiable Learning strategy to explore optimal reasoning paths from the latent probability space so that the model learns to solve intermediate reasoning processes without expert annotations. We further design a Dynamic Adaptive Reasoner to enhance the generalization of unseen questions. Our method achieves 17% improvements in F1-score against BreakRC and shows better interpretability. We take a step forward in building interpretable reasoning methods.

Structural probing work has found evidence for latent syntactic information in pre-trained language models. However, much of this analysis has focused on monolingual models, and analyses of multilingual models have employed correlational methods that are confounded by the choice of probing tasks. In this study, we causally probe multilingual language models (XGLM and multilingual BERT) as well as monolingual BERT-based models across various languages; we do this by performing counterfactual perturbations on neuron activations and observing the effect on models’ subject-verb agreement probabilities. We observe where in the model and to what extent syntactic agreement is encoded in each language. We find significant neuron overlap across languages in autoregressive multilingual language models, but not masked language models. We also find two distinct layer-wise effect patterns and two distinct sets of neurons used for syntactic agreement, depending on whether the subject and verb are separated by other tokens. Finally, we find that behavioral analyses of language models are likely underestimating how sensitive masked language models are to syntactic information.

Complex Question Understanding (CQU) parses complex questions to Question Decomposition Meaning Representation (QDMR) which is a sequence of atomic operators. Existing works are based on end-to-end neural models which do not explicitly model the intermediate states and lack interpretability for the parsing process. Besides, they predict QDMR in a mismatched granularity and do not model the step-wise information which is an essential characteristic of QDMR. To alleviate the issues, we treat QDMR as a computational graph and propose a transition-based method where a decider predicts a sequence of actions to build the graph node-by-node. In this way, the partial graph at each step enables better representation of the intermediate states and better interpretability. At each step, the decider encodes the intermediate state with specially designed encoders and predicts several candidates of the next action and its confidence. For inference, a searcher seeks the optimal graph based on the predictions of the decider to alleviate the error propagation. Experimental results demonstrate the parsing accuracy of our method against several strong baselines. Moreover, our method has transparent and human-readable intermediate results, showing improved interpretability.

Automatic readability assessment (ARA) aims at classifying the readability level of a passage automatically. In the past, manually selected linguistic features are used to classify the passages. However, as the use of deep neural network surges, there is less work focusing on these linguistic features. Recently, many works integrate linguistic features with pre-trained language model (PLM) to make up for the information that PLMs are not good at capturing. Despite their initial success, insufficient analysis of the long passage characteristic of ARA has been done before. To further investigate the promotion of linguistic features on PLMs in ARA from the perspective of passage length, with commonly used linguistic features and abundant experiments, we find that: (1) Linguistic features promote PLMs in ARA mainly on long passages. (2) The promotion of the features on PLMs becomes less significant when the dataset size exceeds 750 passages. (3) By analyzing commonly used ARA datasets, we find Newsela is actually not suitable for ARA. Our code is available at https://github.com/recorderhou/linguistic-features-in-ARA.