Keyword mnemonics are memorable explanations that link new terms to simpler keywords. Prior works generate mnemonics for students, but they do not guide models toward mnemonics students prefer and aid learning. We build SMART, a mnemonic generator trained on feedback from real students learning new terms. To train SMART, we first fine-tune LLaMA-2 on a curated set of user-written mnemonics. We then use LLM alignment to enhance SMART: we deploy mnemonics generated by SMART in a flashcard app to find preferences on mnemonics students favor. We gather 2684 preferences from 45 students across two types: expressed (inferred from ratings) and observed (inferred from student learning), yielding three key findings. First, expressed and observed preferences disagree; what students think is helpful does not fully capture what is truly helpful. Second, Bayesian models can synthesize complementary data from multiple preference types into a single effectiveness signal. SMART is tuned via Direct Preference Optimization on this signal, which we show resolves ties and missing labels in the typical method of pairwise comparisons, augmenting data for LLM output quality gains. Third, mnemonic experts assess SMART as matching GPT-4, at much lower deployment costs, showing the utility of capturing diverse student feedback to align LLMs in education.

Detecting structure in data is the first step to arrive at meaningful representations for systems. This is particularly challenging for dislocation networks evolving as a consequence of plastic deformation of crystalline systems. Our study employs Isomap, a manifold learning technique, to unveil the intrinsic structure of high-dimensional density field data of dislocation structures from different compression axis. The resulting maps provide a systematic framework for quantitatively comparing dislocation structures, offering unique fingerprints based on density fields. Our novel, unbiased approach contributes to the quantitative classification of dislocation structures which can be systematically extended.

The reward system is one of the fundamental drivers of animal behaviors and is critical for survival and reproduction. Despite its importance, the problem of how the reward system has evolved is underexplored. In this paper, we try to replicate the evolution of biologically plausible reward functions and investigate how environmental conditions affect evolved rewards' shape. For this purpose, we developed a population-based decentralized evolutionary simulation framework, where agents maintain their energy level to live longer and produce more children. Each agent inherits its reward function from its parent subject to mutation and learns to get rewards via reinforcement learning throughout its lifetime. Our results show that biologically reasonable positive rewards for food acquisition and negative rewards for motor action can evolve from randomly initialized ones. However, we also find that the rewards for motor action diverge into two modes: largely positive and slightly negative. The emergence of positive motor action rewards is surprising because it can make agents too active and inefficient in foraging. In environments with poor and poisonous foods, the evolution of rewards for less important foods tends to be unstable, while rewards for normal foods are still stable. These results demonstrate the usefulness of our simulation environment and energy-dependent birth and death model for further studies of the origin of reward systems.

Evaluating LLMs and text-to-image models is a computationally intensive task often overlooked. Efficient evaluation is crucial for understanding the diverse capabilities of these models and enabling comparisons across a growing number of new models and benchmarks. To address this, we introduce SubLIME, a data-efficient evaluation framework that employs adaptive sampling techniques, such as clustering and quality-based methods, to create representative subsets of benchmarks. Our approach ensures statistically aligned model rankings compared to full datasets, evidenced by high Pearson correlation coefficients. Empirical analysis across six NLP benchmarks reveals that: (1) quality-based sampling consistently achieves strong correlations (0.85 to 0.95) with full datasets at a 10\% sampling rate such as Quality SE and Quality CPD (2) clustering methods excel in specific benchmarks such as MMLU (3) no single method universally outperforms others across all metrics. Extending this framework, we leverage the HEIM leaderboard to cover 25 text-to-image models on 17 different benchmarks. SubLIME dynamically selects the optimal technique for each benchmark, significantly reducing evaluation costs while preserving ranking integrity and score distribution. Notably, a minimal sampling rate of 1% proves effective for benchmarks like MMLU. Additionally, we demonstrate that employing difficulty-based sampling to target more challenging benchmark segments enhances model differentiation with broader score distributions. We also combine semantic search, tool use, and GPT-4 review to identify redundancy across benchmarks within specific LLM categories, such as coding benchmarks. This allows us to further reduce the number of samples needed to maintain targeted rank preservation. Overall, SubLIME offers a versatile and cost-effective solution for the robust evaluation of LLMs and text-to-image models.

The positive impact of cooperative bots on cooperation within evolutionary game theory is well documented; however, existing studies have predominantly used discrete strategic frameworks, focusing on deterministic actions with a fixed probability of one. This paper extends the investigation to continuous and mixed strategic approaches. Continuous strategies employ intermediate probabilities to convey varying degrees of cooperation and focus on expected payoffs. In contrast, mixed strategies calculate immediate payoffs from actions chosen at a given moment within these probabilities. Using the prisoner's dilemma game, this study examines the effects of cooperative bots on human cooperation within hybrid populations of human players and simple bots, across both well-mixed and structured populations. Our findings reveal that cooperative bots significantly enhance cooperation in both population types across these strategic approaches under weak imitation scenarios, where players are less concerned with material gains. However, under strong imitation scenarios, while cooperative bots do not alter the defective equilibrium in well-mixed populations, they have varied impacts in structured populations across these strategic approaches. Specifically, they disrupt cooperation under discrete and continuous strategies but facilitate it under mixed strategies. These results highlight the nuanced effects of cooperative bots within different strategic frameworks and underscore the need for careful deployment, as their effectiveness is highly sensitive to how humans update their actions and their chosen strategic approach.

We evaluate the effectiveness of GPT-4 Turbo in generating educational questions from NCERT textbooks in zero-shot mode. Our study highlights GPT-4 Turbo's ability to generate questions that require higher-order thinking skills, especially at the "understanding" level according to Bloom's Revised Taxonomy. While we find a notable consistency between questions generated by GPT-4 Turbo and those assessed by humans in terms of complexity, there are occasional differences. Our evaluation also uncovers variations in how humans and machines evaluate question quality, with a trend inversely related to Bloom's Revised Taxonomy levels. These findings suggest that while GPT-4 Turbo is a promising tool for educational question generation, its efficacy varies across different cognitive levels, indicating a need for further refinement to fully meet educational standards.

We introduce SpreadsheetBench, a challenging spreadsheet manipulation benchmark exclusively derived from real-world scenarios, designed to immerse current large language models (LLMs) in the actual workflow of spreadsheet users. Unlike existing benchmarks that rely on synthesized queries and simplified spreadsheet files, SpreadsheetBench is built from 912 real questions gathered from online Excel forums, which reflect the intricate needs of users. The associated spreadsheets from the forums contain a variety of tabular data such as multiple tables, non-standard relational tables, and abundant non-textual elements. Furthermore, we propose a more reliable evaluation metric akin to online judge platforms, where multiple spreadsheet files are created as test cases for each instruction, ensuring the evaluation of robust solutions capable of handling spreadsheets with varying values. Our comprehensive evaluation of various LLMs under both single-round and multi-round inference settings reveals a substantial gap between the state-of-the-art (SOTA) models and human performance, highlighting the benchmark's difficulty.

Transformers can under some circumstances generalize to novel problem instances whose constituent parts might have been encountered during training but whose compositions have not. What mechanisms underlie this ability for compositional generalization? By reformulating multi-head attention as a hypernetwork, we reveal that a low-dimensional latent code specifies key-query specific operations. We find empirically that this latent code is highly structured, capturing information about the subtasks performed by the network. Using the framework of attention as a hypernetwork we further propose a simple modification of multi-head linear attention that strengthens the ability for compositional generalization on a range of abstract reasoning tasks. In particular, we introduce a symbolic version of the Raven Progressive Matrices human intelligence test on which we demonstrate how scaling model size and data enables compositional generalization and gives rise to a functionally structured latent code in the transformer.

Neural Machine Translation models are extremely data and compute-hungry. However, not all data points contribute equally to model training and generalization. Data pruning to remove the low-value data points has the benefit of drastically reducing the compute budget without significant drop in model performance. In this paper, we propose a new data pruning technique: Checkpoints Across Time (CAT), that leverages early model training dynamics to identify the most relevant data points for model performance. We benchmark CAT against several data pruning techniques including COMET-QE, LASER and LaBSE. We find that CAT outperforms the benchmarks on Indo-European languages on multiple test sets. When applied to English-German, English-French and English-Swahili translation tasks, CAT achieves comparable performance to using the full dataset, while pruning up to 50% of training data. We inspect the data points that CAT selects and find that it tends to favour longer sentences and sentences with unique or rare words.

We study the question: How can we select the right data for fine-tuning to a specific task? We call this data selection problem active fine-tuning and show that it is an instance of transductive active learning, a novel generalization of classical active learning. We propose ITL, short for information-based transductive learning, an approach which samples adaptively to maximize information gained about the specified task. We are the first to show, under general regularity assumptions, that such decision rules converge uniformly to the smallest possible uncertainty obtainable from the accessible data. We apply ITL to the few-shot fine-tuning of large neural networks and show that fine-tuning with ITL learns the task with significantly fewer examples than the state-of-the-art.