Controllable learning (CL) emerges as a critical component in trustworthy machine learning, ensuring that learners meet predefined targets and can adaptively adjust without retraining according to the changes in those targets. We provide a formal definition of CL, and discuss its applications in information retrieval (IR) where information needs are often complex and dynamic. The survey categorizes CL according to who controls (users or platforms), what is controllable (e.g., retrieval objectives, users' historical behaviors, controllable environmental adaptation), how control is implemented (e.g., rule-based method, Pareto optimization, Hypernetwork), and where to implement control (e.g.,pre-processing, in-processing, post-processing methods). Then, we identify challenges faced by CL across training, evaluation, task setting, and deployment in online environments. Additionally, we outline promising directions for CL in theoretical analysis, efficient computation, empowering large language models, application scenarios and evaluation frameworks in IR.

Alex Galvagni, CEO of Age of Learning and a former artificial intelligence researcher with NASA, says advances in AI now make it possible to deliver to children "a personalized and supportive" experience in education. ROBOTIC POWER WEAR: A groundbreaking AI-powered exoskeleton developed by researchers at North Carolina State University and the University of North Carolina at Chapel Hill promises to be a game-changer for individuals with mobility issues. ELECTION SEASON: Google on Monday announced that it will have a mandatory requirement for advertisers to disclose election ads that use digitally altered content in depictions of real or realistic-looking people or events. Victor Miller is running for mayor of Cheyenne as AI bot'VIC' (Fox News Digital) 'AI FOR MAYOR': A Wyoming man who filed for the state capital's mayor's race as an AI bot named "VIC" spoke to Fox News Digital this week about VIC's landmark candidacy and a breaking setback he encountered moments before taping. SAFEGUARD SUMMER SOJOURNS: A new study by online protection company McAfee has identified the top five destinations most frequently targeted by cybercriminals for online booking scams.

The school district said it dropped its dealings with AllHere, the company that created "Ed," the sun-shaped chatbot, after "we were notified of their financial collapse." AllHere did not respond to an inquiry this week from The Times and the level of its operation is unclear. In a separate development, a major data breach has affected a data cloud company called Snowflake, which has worked with L.A. Unified. The district said Tuesday that there is no connection to the AllHere situation, and that it is working with investigative agencies to assess the damage and which district records were obtained through a third-party contractor. Meanwhile, the district unplugged the chatbot -- for which AllHere had been paid 3 million -- on June 14, less than three months after unveiling the animated figure as an easy-to-use, conversational companion for students and a soon-to-be-indispensable guide for parents.

Hallucinations -- the generation of untrue claims -- pose a challenge to the application of large language models (LLMs) [1] thereby motivating the development of metrics to evaluate factual precision. We observe that popular metrics using the Decompose-Then-Verify framework, such as FActScore [2], can be manipulated by adding obvious or repetitive claims to artificially inflate scores. We expand the FActScore dataset to design and analyze factual precision metrics, demonstrating that models can be trained to achieve high scores under existing metrics through exploiting the issues we identify. This motivates our new customizable plug-and-play subclaim selection component called Core, which filters down individual subclaims according to their uniqueness and informativeness. Metrics augmented by Core are substantially more robust as shown in head-to-head comparisons. We release an evaluation framework supporting the modular use of Core (https://github.com/zipJiang/Core) and various decomposition strategies, and we suggest its adoption by the LLM community. [1] Hong et al., "The Hallucinations Leaderboard -- An Open Effort to Measure Hallucinations in Large Language Models", arXiv:2404.05904v2 [cs.CL]. [2] Min et al., "FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation", arXiv:2305.14251v2 [cs.CL].

Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using preference datasets, enabling the LLM to generate outputs that align with human preferences. Given the sensitive nature of these preference datasets held by various clients, there is a need to implement RLHF within a federated learning (FL) framework, where clients are reluctant to share their data due to privacy concerns. To address this, we introduce a feasible framework in which clients collaboratively train a binary selector with their preference datasets using our proposed FedBis. With a well-trained selector, we can further enhance the LLM that generates human-preferred completions. Meanwhile, we propose a novel algorithm, FedBiscuit, that trains multiple selectors by organizing clients into balanced and disjoint clusters based on their preferences. Compared to the FedBis, FedBiscuit demonstrates superior performance in simulating human preferences for pairwise completions. Our extensive experiments on federated human preference datasets -- marking the first benchmark to address heterogeneous data partitioning among clients -- demonstrate that FedBiscuit outperforms FedBis and even surpasses traditional centralized training.

Synthetic data is becoming increasingly important for accelerating the development of language models, both large and small. Despite several successful use cases, researchers also raised concerns around model collapse and drawbacks of imitating other models. This discrepancy can be attributed to the fact that synthetic data varies in quality and diversity. Effective use of synthetic data usually requires significant human effort in curating the data. We focus on using synthetic data for post-training, specifically creating data by powerful models to teach a new skill or behavior to another model, we refer to this setting as Generative Teaching. We introduce AgentInstruct, an extensible agentic framework for automatically creating large amounts of diverse and high-quality synthetic data. AgentInstruct can create both the prompts and responses, using only raw data sources like text documents and code files as seeds. We demonstrate the utility of AgentInstruct by creating a post training dataset of 25M pairs to teach language models different skills, such as text editing, creative writing, tool usage, coding, reading comprehension, etc. The dataset can be used for instruction tuning of any base model. We post-train Mistral-7b with the data. When comparing the resulting model Orca-3 to Mistral-7b-Instruct (which uses the same base model), we observe significant improvements across many benchmarks. For example, 40% improvement on AGIEval, 19% improvement on MMLU, 54% improvement on GSM8K, 38% improvement on BBH and 45% improvement on AlpacaEval. Additionally, it consistently outperforms other models such as LLAMA-8B-instruct and GPT-3.5-turbo.

Recently, the demand for psychological counseling has significantly increased as more individuals express concerns about their mental health. This surge has accelerated efforts to improve the accessibility of counseling by using large language models (LLMs) as counselors. To ensure client privacy, training open-source LLMs faces a key challenge: the absence of realistic counseling datasets. To address this, we introduce Cactus, a multi-turn dialogue dataset that emulates real-life interactions using the goal-oriented and structured approach of Cognitive Behavioral Therapy (CBT). We create a diverse and realistic dataset by designing clients with varied, specific personas, and having counselors systematically apply CBT techniques in their interactions. To assess the quality of our data, we benchmark against established psychological criteria used to evaluate real counseling sessions, ensuring alignment with expert evaluations. Experimental results demonstrate that Camel, a model trained with Cactus, outperforms other models in counseling skills, highlighting its effectiveness and potential as a counseling agent. We make our data, model, and code publicly available.

The ability to perform complex tasks from detailed instructions is a key to many remarkable achievements of our species. As humans, we are not only capable of performing a wide variety of tasks but also very complex ones that may entail hundreds or thousands of steps to complete. Large language models and their more recent multimodal counterparts that integrate textual and visual inputs have achieved unprecedented success in performing complex tasks. Yet, most existing benchmarks are largely confined to single-modality inputs (either text or vision), narrowing the scope of multimodal assessments, particularly for instruction-following in multimodal contexts. To bridge this gap, we introduce the instructed-Virtual VISual Decision Making (iWISDM) environment engineered to generate a limitless array of vision-language tasks of varying complexity. Using iWISDM, we compiled three distinct benchmarks of instruction following visual tasks across varying complexity levels and evaluated several newly developed multimodal models on these benchmarks. Our findings establish iWISDM as a robust benchmark for assessing the instructional adherence of both existing and emergent multimodal models and highlight a large gap between these models' ability to precisely follow instructions with that of humans. A typical day in most people's lives involves hundreds or thousands of tasks. Most of which are performed without explicit attention. Just in between getting up and getting to work, one may have already performed 5-15 tasks (taking a shower, shaving, making coffee, getting dressed, etc.). Teaching artificial agents to perform similar seemingly mundane tasks has proven to be an extremely difficult computational problem (Konar, 2018). The challenge becomes more apparent when one realizes that each of these seemingly mundane tasks such as making coffee involves tens of steps/actions (Figure 1). The challenge becomes increasingly more significant once we consider more complex tasks such as operating a device or assembling a piece of furniture from its instruction manual. And yet, these tasks are performed proficiently by most individuals in most situations. Large Language Models (LLMs) have become increasingly capable of comprehending natural language across wide topics and contexts, allowing them to hold meaningful conversations, give expert advice, and analyze data among other features (Brown et al., 2020; Ouyang et al., 2022; Radford et al., 2019). In the meantime, their multimodal counterparts are starting to emerge, signalling broader application of such models across industries. Large Multimodal Models (LMMs) are generally capable of receiving and responding in a range of possible modalities including visual, text, and audio (Alayrac et al., 2022; Liu et al., 2023b; Achiam et al., 2023). These authors contributed equally to this work.

Recently, various encoder-only and encoder-decoder pre-trained models like BERT and T5 have been applied to automatic essay scoring (AES) as small language models. However, existing studies have primarily treated this task akin to a classification problem, focusing solely on outputting scores in the target text without offering interpretations for the generated scores. Departing from the approaches, we introduce Reasoning Distillation-Based Evaluation (RDBE), which integrates interpretability to elucidate the rationale behind model scores while enhancing performance through initial reasoning. This interpretive capability is acquired during training by leveraging generated reasoning from a large language model (LLM) to distill a small language model (SLM). Our experimental results demonstrate the efficacy of RDBE across all scoring rubrics considered in the dataset. RDBE outperforms both zero-shot LLM generation and generation from a baseline fine-tuned model, establishing itself as state-of-the-art in the corresponding dataset. This highlights its practical interpretative output and enhanced performance.

The granular-ball (GB)-based classifier introduced by Xia, exhibits adaptability in creating coarse-grained information granules for input, thereby enhancing its generality and flexibility. Nevertheless, the current GB-based classifiers rigidly assign a specific class label to each data instance and lacks of the necessary strategies to address uncertain instances. These far-fetched certain classification approachs toward uncertain instances may suffer considerable risks. To solve this problem, we construct a robust three-way classifier with shadowed GBs for uncertain data. Firstly, combine with information entropy, we propose an enhanced GB generation method with the principle of justifiable granularity. Subsequently, based on minimum uncertainty, a shadowed mapping is utilized to partition a GB into Core region, Important region and Unessential region. Based on the constructed shadowed GBs, we establish a three-way classifier to categorize data instances into certain classes and uncertain case. Finally, extensive comparative experiments are conducted with 2 three-way classifiers, 3 state-of-the-art GB-based classifiers, and 3 classical machine learning classifiers on 12 public benchmark datasets. The results show that our model demonstrates robustness in managing uncertain data and effectively mitigates classification risks. Furthermore, our model almost outperforms the other comparison methods in both effectiveness and efficiency.