Vision-Language Models (VLMs) have transformed tasks requiring visual and reasoning abilities, such as image retrieval and Visual Question Answering (VQA). Despite their success, VLMs face significant challenges with tasks involving geometric reasoning, algebraic problem-solving, and counting. These limitations stem from difficulties effectively integrating multiple modalities and accurately interpreting geometry-related tasks. Various works claim that introducing a captioning pipeline before VQA tasks enhances performance. We incorporated this pipeline for tasks involving geometry, algebra, and counting. We found that captioning results are not generalizable, specifically with larger VLMs primarily trained on downstream QnA tasks showing random performance on math-related challenges. However, we present a promising alternative: task-based prompting, enriching the prompt with task-specific guidance. This approach shows promise and proves more effective than direct captioning methods for math-heavy problems.

A vast amount of textual data is added to the internet daily, making utilization and interpretation of such data difficult and cumbersome. As a result, automatic text summarization is crucial for extracting relevant information, saving precious reading time. Although many transformer-based models excel in summarization, they are constrained by their input size, preventing them from processing texts longer than their context size. This study introduces three novel algorithms that allow any LLM to efficiently overcome its input size limitation, effectively utilizing its full potential without any architectural modifications. We test our algorithms on texts with more than 70,000 words, and our experiments show a significant increase in BERTScore with competitive ROUGE scores.

English-centric large language models (LLMs) often show strong multilingual capabilities. However, the multilingual performance of these models remains unclear and is not thoroughly evaluated for many languages. Most benchmarks for multilinguality focus on classic NLP tasks, or cover a minimal number of languages. We introduce MEXA, a method for assessing the multilingual capabilities of pre-trained English-centric LLMs using parallel sentences, which are available for more languages than existing downstream tasks. MEXA leverages the fact that English-centric LLMs use English as a kind of pivot language in their intermediate layers. It computes the alignment between English and non-English languages using parallel sentences to evaluate the transfer of language understanding from English to other languages. This alignment can be used to estimate model performance in other languages. We conduct studies using various parallel datasets (FLORES-200 and Bible), models (Llama family, Gemma family, Mistral, and OLMo), and established downstream tasks (Belebele, m-MMLU, and m-ARC). We explore different methods to compute embeddings in decoder-only models. Our results show that MEXA, in its default settings, achieves a statistically significant average Pearson correlation of 0.90 with three established downstream tasks across nine models and two parallel datasets. This suggests that MEXA is a reliable method for estimating the multilingual capabilities of English-centric LLMs, providing a clearer understanding of their multilingual potential and the inner workings of LLMs. Leaderboard: https://huggingface.co/spaces/cis-lmu/Mexa, Code: https://github.com/cisnlp/Mexa.

Abstract-- In this paper, we explore the application of the Decision Transformer, a decision-making algorithm based on the Generative Pre-trained Transformer (GPT) architecture, to multi-vehicle coordination at unsignalized intersections. We formulate the coordination problem so as to find the optimal trajectories for multiple vehicles at intersections, modeling it as a sequence prediction task to fully leverage the power of GPTs as a sequence model. Through extensive experiments, we compare our approach to a reservation-based intersection management system. Our results show that the Decision Transformer can outperform the training data in terms of total travel time and can be generalized effectively to various scenarios, including noise-induced velocity variations, continuous interaction environments, and different vehicle numbers and road configurations. Intelligent traffic management has been shown to substantially minimize delays, enhance safety, and improve overall mobility.

In psychotherapy, therapeutic outcome assessment, or treatment outcome evaluation, is essential for enhancing mental health care by systematically evaluating therapeutic processes and outcomes. Existing large language model approaches often focus on therapist-centered, single-session evaluations, neglecting the client's subjective experience and longitudinal progress across multiple sessions. To address these limitations, we propose IPAEval, a client-Informed Psychological Assessment-based Evaluation framework that automates treatment outcome evaluations from the client's perspective using clinical interviews. IPAEval integrates cross-session client-contextual assessment and session-focused client-dynamics assessment to provide a comprehensive understanding of therapeutic progress. Experiments on our newly developed TheraPhase dataset demonstrate that IPAEval effectively tracks symptom severity and treatment outcomes over multiple sessions, outperforming previous single-session models and validating the benefits of items-aware reasoning mechanisms. In psychotherapy, therapeutic outcome assessment, a.k.a treatment outcome evaluation under clinical settings, refers to the systematic evaluation of therapeutic processes and outcomes (Groth-Marnat, 2009), focusing on factors such as therapist effectiveness (Johns et al., 2019) and treatment efficacy (Jensen-Doss et al., 2018) to improve mental health care delivery. It plays a significant role in enhancing the quality and effectiveness of mental health care by providing actionable insights that guide therapists in refining their treatment approaches (Wampold & Imel, 2015), ultimately leading to better client outcomes and improved therapeutic relationships in real-world clinical practice (Maruish & Leahy, 2000). Over the last couple of years, the emergence of large language models has demonstrated their effectiveness in automatic evaluations, showing a high degree of alignment with human judgment when provided with proper instruction and contextual guidance (Liu et al., 2023; Li et al., 2024b; Kim et al., 2024). This aligns with the "LLMs-as-a-judge" paradigm, where LLMs are employed to simulate human evaluators by providing assessments based on natural language input (Zheng et al., 2023; Wang et al., 2024b). This paradigm has been extended to therapeutic outcome assessment by harnessing LLMs' ability to model complex therapeutic procedures and interactions, offering a novel pathway for automating the assessment of therapeutic efficacy (Chiu et al., 2024; Lee et al., 2024; Li et al., 2024a).

In sensitive domains, such as legal or medial domains, the correctness of information given to users is critical. To address this, the recently introduced task Conversational Tree Search (CTS) provides a graph-based framework for controllable task-oriented dialog in sensitive domains. However, a big drawback of state-of-the-art CTS agents is their long training time, which is especially problematic as a new agent must be trained every time the associated domain graph is updated. The goal of this paper is to eliminate the need for training CTS agents altogether. To achieve this, we implement a novel LLM-based method for zero-shot, controllable CTS agents. We show that these agents significantly outperform state-of-the-art CTS agents (p<0.0001; Barnard Exact test) in simulation. This generalizes to all available CTS domains. Finally, we perform user evaluation to test the agent performance in the wild, showing that our policy significantly (p<0.05; Barnard Exact) improves task-success compared to the state-of-the-art Reinforcement Learning-based CTS agent.

During the pretraining phase, large language models (LLMs) acquire vast amounts of knowledge from extensive text corpora. Nevertheless, in later stages such as fine-tuning and inference, the model may encounter knowledge not covered in the initial training, which can lead to hallucinations and degraded performance. This issue has a profound impact on the model's capabilities, as it will inevitably face out-of-scope knowledge after pretraining. Furthermore, fine-tuning is often required to adapt LLMs to domain-specific tasks. However, this phenomenon limits the model's ability to learn and integrate new information during fine-tuning. The effectiveness of fine-tuning largely depends on the type of knowledge involved. Existing research suggests that fine-tuning the model on partially mastered knowledge-for instance, question-answer pairs where the model has a chance of providing correct responses under non-greedy decoding-can enable the model to acquire new knowledge while mitigating hallucination. Notably, this approach can still lead to the forgetting of fully mastered knowledge, constraining the fine-tuning dataset to a narrower range and limiting the model's overall potential for improvement. Given the model's intrinsic reasoning abilities and the interconnectedness of different knowledge areas, it is likely that as the model's capacity to utilize existing knowledge improves during fine-tuning, previously unmastered knowledge may become more understandable. To explore this hypothesis, we conducted experiments and, based on the results, proposed a two-stage fine-tuning strategy. This approach not only improves the model's overall test accuracy and knowledge retention but also preserves its accuracy on previously mastered content. When fine-tuning on the WikiQA dataset, our method increases the amount of knowledge acquired by the model in this stage by 24%.

Recent Retrieval Augmented Generation (RAG) aims to enhance Large Language Models (LLMs) by incorporating extensive knowledge retrieved from external sources. However, such approach encounters some challenges: Firstly, the original queries may not be suitable for precise retrieval, resulting in erroneous contextual knowledge; Secondly, the language model can easily generate inconsistent answer with external references due to their knowledge boundary limitation. To address these issues, we propose the chain-of-verification (CoV-RAG) to enhance the external retrieval correctness and internal generation consistency. Specifically, we integrate the verification module into the RAG, engaging in scoring, judgment, and rewriting. To correct external retrieval errors, CoV-RAG retrieves new knowledge using a revised query. To correct internal generation errors, we unify QA and verification tasks with a Chain-of-Thought (CoT) reasoning during training. Our comprehensive experiments across various LLMs demonstrate the effectiveness and adaptability compared with other strong baselines. Especially, our CoV-RAG can significantly surpass the state-of-the-art baselines using different LLM backbones.

While large code language models have made significant strides in AI-assisted coding tasks, there are growing concerns about privacy challenges. The user code is transparent to the cloud LLM service provider, inducing risks of unauthorized training, reading, and execution of the user code. In this paper, we propose CodeCipher, a novel method that perturbs privacy from code while preserving the original response from LLMs. CodeCipher transforms the LLM's embedding matrix so that each row corresponds to a different word in the original matrix, forming a token-to-token confusion mapping for obfuscating source code. The new embedding matrix is optimized by minimizing the task-specific loss function. To tackle the challenge of the discrete and sparse nature of word vector spaces, CodeCipher adopts a discrete optimization strategy that aligns the updated vector to the nearest valid token in the vocabulary before each gradient update. We demonstrate the effectiveness of our approach on three AI-assisted coding tasks including code completion, summarization, and translation. Results show that our model successfully confuses the privacy in source code while preserving the original LLM's performance.

Dense high dimensional vectors are becoming increasingly vital in fields such as computer vision, machine learning, and large language models (LLMs), serving as standard representations for multimodal data. Now the dimensionality of these vector can exceed several thousands easily. Despite the nearest neighbor search (NNS) over these dense high dimensional vectors have been widely used for retrieval augmented generation (RAG) and many other applications, the effectiveness of NNS in such a high-dimensional space remains uncertain, given the possible challenge caused by the "curse of dimensionality." To address above question, in this paper, we conduct extensive NNS studies with different distance functions, such as $L_1$ distance, $L_2$ distance and angular-distance, across diverse embedding datasets, of varied types, dimensionality and modality. Our aim is to investigate factors influencing the meaningfulness of NNS. Our experiments reveal that high-dimensional text embeddings exhibit increased resilience as dimensionality rises to higher levels when compared to random vectors. This resilience suggests that text embeddings are less affected to the "curse of dimensionality," resulting in more meaningful NNS outcomes for practical use. Additionally, the choice of distance function has minimal impact on the relevance of NNS. Our study shows the effectiveness of the embedding-based data representation method and can offer opportunity for further optimization of dense vector-related applications.