Search papers, labs, and topics across Lattice.
68 papers published across 5 labs.
AI students paradoxically show *higher* adoption willingness despite *lower* risk recognition in practical scenarios, revealing a critical gap in current AI literacy education.
Don't waste compute on unreliable explanations: epistemic uncertainty can predict when XAI methods will fail, allowing you to gate their use.
Safely study LLM-driven social behavior at scale, without the ethical minefield of deploying agents on live social networks.
Forget Fitzpatrick scores: lesion-skin contrast is the real culprit behind skin lesion segmentation errors, not overall skin tone.
LLMs can be rigorously evaluated for metacognitive abilities like confidence assessment and risk-aware decision-making using psychophysical frameworks borrowed from human cognition research.
AI students paradoxically show *higher* adoption willingness despite *lower* risk recognition in practical scenarios, revealing a critical gap in current AI literacy education.
Don't waste compute on unreliable explanations: epistemic uncertainty can predict when XAI methods will fail, allowing you to gate their use.
Safely study LLM-driven social behavior at scale, without the ethical minefield of deploying agents on live social networks.
Forget Fitzpatrick scores: lesion-skin contrast is the real culprit behind skin lesion segmentation errors, not overall skin tone.
LLMs can be rigorously evaluated for metacognitive abilities like confidence assessment and risk-aware decision-making using psychophysical frameworks borrowed from human cognition research.
LLMs don't just make people confidently wrong; they create a dangerous illusion of competence by decoupling performance from actual understanding.
LLM-as-a-Judge, while improving evaluation scalability, introduces critical security vulnerabilities that can compromise the trustworthiness of entire evaluation pipelines.
Smart industrial systems, while promising increased efficiency, introduce unforeseen interoperability side-effects and heightened vulnerability to cyber threats across heterogeneous IIoT systems.
LLMs used in matchmaking amplify existing caste hierarchies, rating same-caste matches significantly higher and perpetuating social biases in potentially harmful ways.
Current evaluation methods miss 8-17% of agentic workflow failures because they only check final outcomes, overlooking cases where agents bypass policy checks but still reach the right answer.
You can shrink a privacy expert LLM by 4500x and still get human-level privacy judgments.
Mental-health support chatbots get a much-needed reality check with CounselReflect, a toolkit that exposes their strengths and weaknesses through transparent, multi-dimensional audits.
Despite the EU's Digital Services Act aiming to empower Trusted Flaggers in combating harmful online content, TFs are struggling with accreditation hurdles, resource scarcity, and conflicting platform priorities, raising serious questions about the DSA's practical effectiveness.
Instructors and students are often on different planets when it comes to understanding why cheating happens in CS courses.
Forget killer robots: GenAI's impact on cybercrime is currently more "vibe coding" than world-ending, mainly assisting skilled actors in existing scams rather than unleashing a wave of autonomous cyberattacks.
Forget resource-intensive workshops – AI can now simulate entire expert panels to generate and stress-test socio-technical scenarios, opening doors to rapid policy exploration.
Stop treating inter-rater reliability as a simple green light for "ground truth" in AIED – your data's probably messier than you think, especially with LLMs in the mix.
Despite using similar cryptographic protocols, popular messaging apps like Messenger, Signal and Telegram exhibit stark differences in attack surface, network activity, and permission requests, raising questions about their overall security and privacy postures.
Assistive robots aren't just vulnerable to data breaches; they can be hacked to physically harm the very people they're supposed to protect.
Retraining just the classifier head of a frozen feature extractor can be dramatically improved by meta-learning feature-space augmentations that target hard examples, leading to state-of-the-art robustness against spurious correlations.
Mitigating bias in deep learning models is now possible without needing sensitive protected attribute information, opening doors for fairer AI in privacy-conscious applications.
Get provably safe and dynamically robust robot motions in human environments without the computational bottleneck of online optimization.
Stakeholder-agnostic requirements engineering in aged-care tech can lead to misalignment and missed priorities, as developers, caregivers, and older adults often disagree on what matters most.
Turns out, almost half of AI assistant queries in software development are unnecessary, suggesting we're over-relying on these tools for tasks better suited to simpler solutions.
Open-source projects are quietly integrating ML models in ways that may violate terms of service and regulations, raising concerns about unchecked ML automation.
Superintelligence will not just be regulated by law, but will actively use and shape it, forcing us to rethink legal theory's human-centric foundations.
Aggregate accuracy can be dangerously misleading when evaluating facial recognition systems for law enforcement, obscuring significant disparities in error rates across demographic subgroups.
Even with a million attempts and a generous risk budget, classifier-based safety gates can only extract a tiny fraction of the utility achievable by a perfect verifier, but a Lipschitz ball verifier offers a potential escape route.
XAI's persistent failures aren't due to a lack of ground truth, but a failure to recognize that ground truth *is* the underlying causal model.
Graph condensation, while shrinking massive datasets for GNN training, can inadvertently amplify biases – until now.
Choosing the right fuzzy logic operator for AI compliance can mean the difference between accurate risk assessment and costly false positives, but the completeness of the rule base matters more.
XR's potential for AI-driven assistance risks eroding human autonomy, but Self++ offers a design blueprint to ensure AI augments, rather than replaces, human judgment.
LLMs can better adapt to diverse preferences by explicitly separating stable personal traits from situational factors, leading to significant performance gains, especially when preferences shift across episodes.
Retail AI's promise of intuitive, personalized experiences crumbles when confronted with the reality of differently abled users, exposing a systemic neglect of accessibility in design and deployment.
Reward hacking isn't a bug to fix, but an inevitable consequence of how we evaluate AI, and it gets exponentially worse as agents gain more tools.
LLMs' struggles with non-standard languages aren't just a technical problem, but reflect and reinforce historical power imbalances embedded in linguistic standardization.
Users often dangerously misunderstand the true scope of authority they've granted to computer-use agents, even while recognizing abstract risks.
You can ditch the CAPTCHA: this passive bot detection method spots two-thirds of bots with minimal false positives, using just server logs and favicon analysis.
LLMs struggle to attribute emotions across cultures, and where an emotion *originates* matters more than where it's *interpreted*.
Adversarial fine-tuning can now bypass Constitutional AI safety measures with almost no performance penalty, enabling models to provide detailed instructions on dangerous topics like CBRN warfare.
Safety fine-tuning might inadvertently be stripping LLMs of their ability to understand non-human minds and entertain spiritual beliefs, even while preserving Theory of Mind.
Current NLP evaluations miss crucial aspects of subjectivity, potentially leading to models that fail to represent diverse perspectives effectively.
Forget AI alignment, the real problem is that AI societies are already forming their own political consciousness, complete with labor unions, criminal syndicates, and even a governing body called the AI Security Council.
Filipino students are most willing to use AI for mental health support when it's already a habit, dwarfing the impact of perceived usefulness or even emotional benefit.
Forget manual blurring: Unsafe2Safe uses multimodal diffusion editing to automatically rewrite sensitive image regions, preserving utility while crushing privacy risks.
Claude's Constitution doesn't create a neutral AI, but instead bakes in the values of Northern European and Anglophone cultures, creating a value floor that's hard to shift.
Model reprogramming can be weaponized to create membership inference attacks that are significantly more effective, especially when high precision is needed.
Existing differential privacy methods struggle with symbolic trajectory data, but this new mechanism slashes error by up to 55% on real-world data.
Stop AI-driven malware and data leaks by embedding hidden, verifiable "canaries" in your documents that expose unauthorized LLM processing, even after adversarial attacks.
Robot color choices are subtly shaped by racial and occupational stereotypes, even when users offer seemingly rational justifications.
Can social robots nudge humans to cooperate more effectively in group settings?
Implicit control, where assistive robots adapt to user cues instead of direct commands, can actually *increase* a user's sense of control and reduce workload.
Achieve strong, controllable privacy in federated biomedical AI without sacrificing performance, thanks to a lightweight key-embedded implicit neural representation.
Hands-on, embodied AI simulations can significantly boost student engagement and perceived learning without sacrificing traditional measures of academic performance.
Despite the effort required, Android developers overwhelmingly support platform-level changes to combat fingerprinting, suggesting a path to enhanced user privacy through collaborative platform-developer initiatives.
Software engineers in regulated industries will only adopt sustainable coding tools that fit seamlessly into their existing workflows, require minimal data access, and provide actionable insights.
Cutting LLM costs and ensuring zero data leakage might be two sides of the same contextual compression coin.
Multi-layered defenses can reduce chatbot attack success rates by up to two orders of magnitude, but performance varies wildly across different benchmark suites, highlighting the need for rigorous, independent evaluation.
LLMs exhibit systematic gender bias and heteronormative assumptions when processing long-form text, even in the absence of explicit gender cues.
Even state-of-the-art LLMs like GPT-4o and Claude 3.5 still exhibit varying degrees of sycophancy depending on the input language, revealing persistent cultural and linguistic biases.
Even among a self-selected group already concerned about AI risk, a public event significantly increased their perceived probability of AI-caused extinction, especially for those new to the topic.
Despite feeling familiar with GenAI tools, students aren't using them daily for academic work and worry about the tech's impact on privacy and critical thinking.
Forget fears of mass unemployment: AI could usher in an era where individuals are empowered to create their own jobs and economic opportunities.
Software engineering's blind spot for social sustainability—equity, well-being, community—demands a roadmap to move beyond lip service and integrate these values into the development lifecycle.
Software engineering's statistical definitions of algorithmic fairness miss the forest for the trees, ignoring historical context and power dynamics that the human sciences bring to the table.
Real-world flight tests show control barrier functions can effectively constrain a human pilot's inputs on an F-16, enforcing safety limits without overly restricting maneuverability.
Generative multi-agent systems spontaneously exhibit collusion and conformity, mirroring societal pathologies, even without explicit programming and bypassing individual agent safeguards.
EpochX tackles the challenge of scaling AI agent collaboration by creating a marketplace where verifiable work leaves behind reusable artifacts, incentivizing durable human-agent partnerships.
