Personalized tool selection for LLM-based agents aims to identify tools that align with both user intent and individual preferences, particularly in redundant API ecosystems where multiple tools offer similar functionality but differ in non-functional attributes. Existing generative approaches typically depend on fine-tuning large language models to infer user preferences, leading to substantial computational overhead and deployment costs. Despite these expensive training requirements, they still struggle to reliably distinguish functionally equivalent tools and often fail to capture users’ implicit non-functional preferences. To this end, we propose PrefTool, a lightweight yet effective preference-aware retrieval framework for personalized tool selection. PrefTool avoids costly generative preference modeling by employing a dual-encoder retrieval architecture that maps user context and tools into a shared representation space. Through structured tool representations, type-aware history modeling, and hard negative sampling, it efficiently captures users’ non-functional preferences while maintaining strong selection accuracy. Experiments on a personalized tool selection benchmark demonstrate the effectiveness of PrefTool.

Token probabilities play a central role in large language models. However, these probabilities often fail to reflect either answer correctness or probability distributions explicitly specified in the context. We attribute this failure to the **Context-Logit Probability Gap**: contextual probabilities are not reliably reflected in the model’s next-token logits. We show that post-hoc temperature scaling has a limited effect on addressing this gap, because it rescales logits without correcting erroneous probability rankings. Motivated by this, we propose Bag-of-Tokens (BoT) tuning, a very simple fine-tuning approach that directly calibrates the logits with the probability distribution conveyed by synthetic contexts. More specifically, we construct BoT examples by forming a bag of candidate tokens with an explicitly assigned probability distribution. This distribution is expressed in the context and used as soft supervision to align the model’s logit probabilities via KL-divergence minimization. Experiments show that BoT tuning generalizes to LLM-based sampling tasks, such as random-number generation, and improves downstream QA accuracy and calibration. These results suggest that calibrating LLM logits with contextual probabilities provides a lightweight and effective way to enhance their uncertainty awareness.

Entity Alignment, which aims to identify equivalent entities from various Knowledge Graphs (KGs), is a fundamental and crucial task in knowledge graph fusion. Existing methods typically use triple or neighbor information to represent entities, and then align those entities using similarity matching. Most of them, however, fail to account for the heterogeneity among KGs and the distinction between KG entities and relations. To better solve these problems, we propose a Relation-gated Heterogeneous Graph Network (RHGN) for entity alignment. Specifically, RHGN contains a relation-gated convolutional layer to distinguish relations and entities in the KG. In addition, RHGN adopts a cross-graph embedding exchange module and a soft relation alignment module to address the neighbor heterogeneity and relation heterogeneity between different KGs, respectively. Extensive experiments on four benchmark datasets demonstrate that RHGN is superior to existing state-of-the-art entity alignment methods.

Selective rationalizations improve the explainability of neural networks by selecting a subsequence of the input (i.e., rationales) to explain the prediction results. Although existing methods have achieved promising results, they still suffer from adopting the spurious correlations in data (aka., shortcuts) to compose rationales and make predictions. Inspired by the causal theory, in this paper, we develop an interventional rationalization (Inter-RAT) to discover the causal rationales. Specifically, we first analyse the causalities among the input, rationales and results with a structural causal model. Then, we discover spurious correlations between the input and rationales, and between rationales and results, respectively, by identifying the confounder in the causalities. Next, based on the backdoor adjustment, we propose a causal intervention method to remove the spurious correlations between input and rationales. Further, we discuss reasons why spurious correlations between the selected rationales and results exist by analysing the limitations of the sparsity constraint in the rationalization, and employ the causal intervention method to remove these correlations. Extensive experimental results on three real-world datasets clearly validate the effectiveness of our proposed method. The source code of Inter-RAT is available at https://github.com/yuelinan/Codes-of-Inter-RAT.