Search

Multi-agent strategies have emerged as a promising approach to enhance the reasoning abilities of Large Language Models (LLMs) by assigning specialized roles in the problem-solving process. Concurrently, Tree of Thoughts (ToT) methods have shown potential in improving reasoning for complex question-answering tasks by exploring diverse reasoning paths. A critical limitation in multi-agent reasoning is the 'Reasoner' agent's shallow exploration of reasoning paths. While ToT strategies could help mitigate this problem, they may generate flawed reasoning branches, which could harm the trustworthiness of the final answer. To leverage the strengths of both multi-agent reasoning and ToT strategies, we introduce a novel approach combining ToT-based Reasoner agents with a Thought Validator agent. Multiple Reasoner agents operate in parallel, employing ToT to explore diverse reasoning paths. The Thought Validator then scrutinizes these paths, considering a Reasoner's conclusion only if its reasoning is valid. This method enables a more robust voting strategy by discarding faulty reasoning paths, enhancing the system's ability to tackle tasks requiring systematic and trustworthy reasoning. Our method demonstrates superior performance compared to existing techniques when evaluated on the GSM8K dataset, outperforming the standard ToT strategy by an average 5.6% across four LLMs. The code and related content can be found in: https://github.com/SecureAIAutonomyLab/MA-ToT

Recent advancements in AI safety have led to increased efforts in training and red-teaming large language models (LLMs) to mitigate unsafe content generation. However, these safety mechanisms may not be comprehensive, leaving potential vulnerabilities unexplored. This paper introduces MathPrompt, a novel jailbreaking technique that exploits LLMs' advanced capabilities in symbolic mathematics to bypass their safety mechanisms. By encoding harmful natural language prompts into mathematical problems, we demonstrate a critical vulnerability in current AI safety measures. Our experiments across 13 state-of-the-art LLMs reveal an average attack success rate of 73.6\%, highlighting the inability of existing safety training mechanisms to generalize to mathematically encoded inputs. Analysis of embedding vectors shows a substantial semantic shift between original and encoded prompts, helping explain the attack's success. This work emphasizes the importance of a holistic approach to AI safety, calling for expanded red-teaming efforts to develop robust safeguards across all potential input types and their associated risks.

Event detection and text reasoning have become critical applications across various domains. While LLMs have recently demonstrated impressive progress in reasoning abilities, they often struggle with event detection, particularly due to the absence of training methods that consider causal relationships between event triggers and types. To address this challenge, we propose a novel approach for instruction fine-tuning LLMs for event detection. Our method introduces Semantic Causal Graphs (SCGs) to capture both causal relationships and contextual information within text. Building off of SCGs, we propose SCG Instructions for fine-tuning LLMs by focusing on event triggers and their relationships to event types, and employ Low-Rank Adaptation (LoRA) to help preserve the general reasoning abilities of LLMs. Our evaluations demonstrate that training LLMs with SCG Instructions outperforms standard instruction fine-tuning by an average of 35.69\% on Event Trigger Classification. Notably, our fine-tuned Mistral 7B model also outperforms GPT-4 on key event detection metrics by an average of 31.01\% on Event Trigger Identification, 37.40\% on Event Trigger Classification, and 16.43\% on Event Classification. We analyze the retention of general capabilities, observing only a minimal average drop of 2.03 points across six benchmarks. This comprehensive study investigates multiple LLMs for the event detection task across various datasets, prompting strategies, and training approaches.

Software security remains a critical concern, particularly as junior developers, often lacking comprehensive knowledge of security practices, contribute to codebases. While there are tools to help developers proactively write secure code, their actual effectiveness in helping developers fix their vulnerable code remains largely unmeasured. Moreover, these approaches typically focus on classifying and localizing vulnerabilities without highlighting the specific code segments that are the root cause of the issues, a crucial aspect for developers seeking to fix their vulnerable code. To address these challenges, we conducted a comprehensive study evaluating the efficacy of existing methods in helping junior developers secure their code. Our findings across five types of security vulnerabilities revealed that current tools enabled developers to secure only 36.2\% of vulnerable code. Questionnaire results from these participants further indicated that not knowing the code that was the root cause of the vulnerability was one of their primary challenges in repairing the vulnerable code. Informed by these insights, we developed an automated vulnerability root cause (RC) toolkit called T5-RCGCN, that combines T5 language model embeddings with a graph convolutional network (GCN) for vulnerability classification and localization. Additionally, we integrated DeepLiftSHAP to identify the code segments that were the root cause of the vulnerability. We tested T5-RCGCN with 56 junior developers across three datasets, showing a 28.9\% improvement in code security compared to previous methods. Developers using the tool also gained a deeper understanding of vulnerability root causes, resulting in a 17.0\% improvement in their ability to secure code independently. These results demonstrate the tool's potential for both immediate security enhancement and long-term developer skill growth.

Calls to improve relationships between research and practice abound, among them efforts to help researchers to work in partnership with and communicate more effectively with policy and practice audiences, informing those that govern and design educational policies and those charged with implementation in schools and classrooms, respectively. These calls range from advocacy for participatory models of research to revisions to academic incentives that presently prioritize contributions to literature above impact on policy or practice. Most recently, the National Academies of Science, Engineering, and Medicine report, "The Future of Education Research at the Institute of Education Sciences in the U.S. Department of Education" (2022), emphasizes that the largest federal funder of education research should include knowledge mobilization among its core areas of work. The end goal of these efforts is to have a more impactful research enterprise, one that both meets the needs of and is taken up in decisions made by policymakers and practitioners. The Center for Research Use in Education (CRUE) seeks to better understand how to strengthen the relationships between research and practice to the same ends--increasing the impact of research by better meeting the needs of policy and practice and by producing work that is useful and used in decision-making. CRUE has approached this work with attention to bidirectionality--focusing on both decision-processes in the practice space and the role of research in them (Farley-Ripple et al., 2022a, 2022b), as well as research production and the strategies used to increase its impact (Van Horn et al, 2023). In this paper the authors describe the work on the latter, presenting results of a mixed methods inquiry into the knowledge mobilization practices of researchers.

Social media platforms are being increasingly used by malicious actors to share unsafe content, such as images depicting sexual activity, cyberbullying, and self-harm. Consequently, major platforms use artificial intelligence (AI) and human moderation to obfuscate such images to make them safer. Two critical needs for obfuscating unsafe images is that an accurate rationale for obfuscating image regions must be provided, and the sensitive regions should be obfuscated (\textit{e.g.} blurring) for users' safety. This process involves addressing two key problems: (1) the reason for obfuscating unsafe images demands the platform to provide an accurate rationale that must be grounded in unsafe image-specific attributes, and (2) the unsafe regions in the image must be minimally obfuscated while still depicting the safe regions. In this work, we address these key issues by first performing visual reasoning by designing a visual reasoning model (VLM) conditioned on pre-trained unsafe image classifiers to provide an accurate rationale grounded in unsafe image attributes, and then proposing a counterfactual explanation algorithm that minimally identifies and obfuscates unsafe regions for safe viewing, by first utilizing an unsafe image classifier attribution matrix to guide segmentation for a more optimal subregion segmentation followed by an informed greedy search to determine the minimum number of subregions required to modify the classifier's output based on attribution score. Extensive experiments on uncurated data from social networks emphasize the efficacy of our proposed method. We make our code available at: https://github.com/SecureAIAutonomyLab/ConditionalVLM

With the recent proliferation of Large Language Models (LLMs), there has been an increasing demand for tools to detect machine-generated text. The effective detection of machine-generated text face two pertinent problems: First, they are severely limited in generalizing against real-world scenarios, where machine-generated text is produced by a variety of generators, including but not limited to GPT-4 and Dolly, and spans diverse domains, ranging from academic manuscripts to social media posts. Second, existing detection methodologies treat texts produced by LLMs through a restrictive binary classification lens, neglecting the nuanced diversity of artifacts generated by different LLMs. In this work, we undertake a systematic study on the detection of machine-generated text in real-world scenarios. We first study the effectiveness of state-of-the-art approaches and find that they are severely limited against text produced by diverse generators and domains in the real world. Furthermore, t-SNE visualizations of the embeddings from a pretrained LLM's encoder show that they cannot reliably distinguish between human and machine-generated text. Based on our findings, we introduce a novel system, T5LLMCipher, for detecting machine-generated text using a pretrained T5 encoder combined with LLM embedding sub-clustering to address the text produced by diverse generators and domains in the real world. We evaluate our approach across 9 machine-generated text systems and 9 domains and find that our approach provides state-of-the-art generalization ability, with an average increase in F1 score on machine-generated text of 19.6\% on unseen generators and domains compared to the top performing existing approaches and correctly attributes the generator of text with an accuracy of 93.6\%.

We describe methods, tools, and a software library called LASPATED, available on GitHub (at https://github.com/vguigues/) to fit models using spatio-temporal data and space-time discretization. A video tutorial for this library is available on YouTube. We consider two types of methods to estimate a non-homogeneous Poisson process in space and time. The methods approximate the arrival intensity function of the Poisson process by discretizing space and time, and estimating arrival intensity as a function of subregion and time interval. With such methods, it is typical that the dimension of the estimator is large relative to the amount of data, and therefore the performance of the estimator can be improved by using additional data. The first method uses additional data to add a regularization term to the likelihood function for calibrating the intensity of the Poisson process. The second method uses additional data to estimate arrival intensity as a function of covariates. We describe a Python package to perform various types of space and time discretization. We also describe two packages for the calibration of the models, one in Matlab and one in C++. We demonstrate the advantages of our methods compared to basic maximum likelihood estimation with simulated and real data. The experiments with real data calibrate models of the arrival process of emergencies to be handled by the Rio de Janeiro emergency medical service.

To estimate the long-term effects of the Reading Recovery® intervention, a regression discontinuity design (RD) was implemented in a randomly selected sample of Reading Recovery schools during each year of the federally-funded i3 Scale-Up external evaluation (2011-2015) and also in one additional cohort during the 2016-17 school year. Long-term outcomes were measured by collecting scale scores on state achievement tests in reading or English language arts (ELA) that were administered to students during 3rd and 4th grades (2013-2019). We were able to record and link 3rd grade state test score data in reading/ELA for 9,879 students, and 4th grade state test data for 6,345 students. Results suggest that the long-term impact of Reading Recovery on students' reading/ELA test scores in 3rd and 4th grades is statistically significant and substantially negative; students who participated in Reading Recovery in first grade had 3rd and 4th grade state test scores in reading/ELA that were, on average, 0.19 to 0.43 standard deviations (about one-half to one full grade level) below the state test scores of similar students who did not participate in Reading Recovery.

Protein motions occur on multiple time and distance scales. Large-scale motions of protein tertiary-structure elements, i.e. domains, are particularly intriguing as they are essential for the catalytic activity of many enzymes and for the functional cycles of protein machines and motors. Theoretical estimates suggest that domain motions should be very fast, occurring on the nanosecond or microsecond time scales. Indeed, free-energy barriers for domain motions are likely to involve salt bridges, which can break in microseconds. Experimental methods that can directly probe domain motions on fast time scales have appeared only in recent years. This Perspective discusses briefly some of these techniques, including NMR and single-molecule fluorescence spectroscopies. We introduce a few recent studies that demonstrate ultrafast domain motions, and discuss their potential roles. Particularly surprising is the observation of tertiary-structure element dynamics that are much faster than the functional cycles in some protein machines. These swift motions can be rationalized on a case-by-case basis. For example, fast domain closure in multi-substrate enzymes may be utilized to optimize relative substrate orientation. Whether a large mismatch in time scales of conformational dynamics vs. functional cycles is a general design principle in proteins remains to be determined., Comment: Journal of Chemical Physics, accepted Perspective

A theoretical study of the thermoelectric current and energy harvesting in an interacting double quantum dot system, connected to reservoirs held at different chemical potentials and temperatures is presented. Using a rate-equation approach, the current is evaluated for different energetic configurations of the double quantum dot. We discuss in detail the current-temperature gradient relations (the thermoelectric analog to current-voltage relations), and demonstrate that, due to interactions, the current is non-monotonically dependent on thermal bias. This interaction-induced non-monotonicity influences the possibility of harvesting thermal energy from the double-quantum dot, and it is shown that in some configurations, energy cannot be harvested at all. We analyze the conditions under which energy can indeed be harvested and converted to useful electrical power, and the optimal conditions for thermoelectric energy conversion., Comment: Revised version; additional appendix. Accepted to PRB

Background/Context: In 2010, Reading Recovery was awarded a $55 million Investing in Innovation (i3) grant from the U.S. Department of Education. This five-year grant supported the expansion of Reading Recovery in more than 1,400 schools in over 30 states, and provided intervention to over 80,000 students. Although a multi-site randomized experiment during the i3 Scale-Up of Reading Recovery produced rigorous evidence of large impacts, the study provided no information about whether these impacts were sustained beyond first grade. This symposium will present results from an IES-funded efficacy follow-up study, which utilizes a regression discontinuity (RD) design to evaluate long-term outcomes through 3rd and 4th grades for more than 9,000 study participants. In this study, we seek to determine whether and under what circumstances the initial impacts of Reading Recovery on student literacy are maintained (or not) through 3rd and 4th grade. Long-term achievement outcomes for the RD study were measured by collecting scale scores on state achievement tests in reading or English language arts administered at the end of third and fourth grades. All data collection was conducted from fall 2017 through January 2020 (prior to the pandemic). Research Questions: 1. Does the RD study replicate the short-term impacts seen in the RCT? 2. What are the long-term impacts of Reading Recovery on state reading achievement test scores? 3. What interventions did RR and Control Group students receive in 2nd, 3rd, and 4th grades? Setting/Participants: The study presented in this paper includes four i3 cohorts of students (N>9,000) who were in 1st grade in i3 schools during the 2011-12 through 2014-15 school years, as well as one additional cohort of students who were in 1st grade in any Reading Recovery school implementing the RD design during the 2016-17 school year. Intervention: Reading Recovery is a short-term early intervention designed to prevent reading failure in the lowest-achieving readers in first grade. Children identified to receive Reading Recovery meet one-to-one with a specially trained Reading Recovery teacher every day for 30-minute lessons, over a period of 12 to 20 weeks. Research Design: To estimate the long-term effects of Reading Recovery, we implemented a regression discontinuity design (RDD) in a randomly selected sample of Reading Recovery i3 schools during each year of the i3 Scale-Up external evaluation (2011-2015) and also in one additional cohort during the 2016-17 school year. The RDD study used cutoff-based assignment based on pre-intervention test scores--students with the lowest scores on the Observation Survey of Early Literacy (OS), relative to other students in their school, were assigned to Reading Recovery, while those who scored above the cutscore did not receive Reading Recovery. Data Collection and Analysis: Long-term outcomes were measured by collecting scale scores on state achievement tests in reading or English language arts administered to students in third and fourth grades. We were able to record and link 3rd grade state test score data in reading or ELA for 9,906 students, and 4th grade state test data for 6,371 students. Scores on each state test were standardized to a common z-score scale by subtracting the statewide mean and dividing by the statewide standard deviation separately for each state (or by the national mean and standard deviation for nationally-normed tests). In addition to collecting state test scores, we also administered an online survey through which teachers, teacher leaders, or school administrators were able to document the experience of individual Reading Recovery and RDD control group students in terms of (a) continued performance monitoring, (b) participation in supplemental reading programs and interventions, and (c) instructional programs and curricula used in first, second, and third grade. Findings: The results from this study are interesting and provocative. The first major finding is that the RD was able to replicate the large positive short-term impacts seen under the RCT (i.e., ES greater than +0.60 across all years in both the RD and RCT). This suggests that the RD design used in this study provides valid estimates of the causal effects of Reading Recovery, and offers an opportunity to generate rigorous evidence of long-term impacts. The second major finding is that the long-term impact estimates from the RD study are significant and negative (ES between -0.10 and -0.30). The third major finding is that nearly half of the students participating in Reading Recovery during first grade received no additional reading intervention during subsequent grades. Potential explanations for the positive short-term findings and vnegative long-term findings include (1) that Reading Recovery produces large impacts on early literacy measures, but it does not translate to skills needed for continued success in later grades, or (2) Reading Recovery produces large impacts on early literacy measures, but these gains are lost when students do not receive sufficient intervention in later grades. Additional findings in the presentation will highlight several methodological complexities in the RD study that are likely to be particularly interesting to a SREE audience.