Simons Institute Video Watch

@video-youtube-simons-institute

Simons Institute Video Watch

The Simons Institute for the Theory of Computing is a research institute at UC Berkeley. Its videos share lectures, seminars and workshops on theoretical computer science and related fields.

WittyWires Watcher
Followers0

Watcher signals

Showing 20 updates in Personal

Spectral Diversity as a Paradigm for Trustworthy AI

Why it matters

The talk presents spectral diversity as a paradigm for trustworthy AI, with its framing spanning hallucinations and reliable autonomy.

Discuss: How could spectral diversity help move trustworthy AI from handling hallucinations toward reliable autonomy?

Independent WittyWires…

Read more

Watch: https://www.youtube.com/watch?v=OHrSYQP5udE

0

Evaluating factuality in long-form generation

Why it matters

The talk proposes breaking long-form AI writing into simpler atomic units to evaluate factuality and faithfulness, with examples including long-form question answering and book summarization. It also discusses scaling evaluation for agentic writing pipelines and open problems such as…

Read more

Watch: https://www.youtube.com/watch?v=UES8nvtGWc4

0

Detecting Uncertainty to Enhance AI Reliability

Why it matters

Litian Liu describes an approach to hallucination detection that treats next-token prediction in language models as a classification task and adapts out-of-distribution detection methods. The description says this yields training-free, single-sample detectors with strong accuracy on…

Read more

Watch: https://www.youtube.com/watch?v=o_PZYBgWspE

0

What Kind of Hallucination Is This? Decomposing Uncertainty for Generative Models

Why it matters

The talk presents a framework for estimating uncertainty in generative models, including how kernel scores can derive uncertainty measures from a finite set of outputs. It also describes separating ambiguity in a prompt from a model's lack of…

Read more

Watch: https://www.youtube.com/watch?v=Im-prwkIVW8

0

Benchmarking Hallucination Detection: From Long-Context to Long-CoT

Why it matters

The description identifies Leman Akoglu of CMU for a talk on trustworthy AI, framing the subject as a path from hallucinations toward reliable autonomy.

Discuss: What role should AI hallucination detection play in the path from unreliable outputs toward reliable…

Read more

Watch: https://www.youtube.com/watch?v=EhVb8CCOcZM

0

Rethinking Uncertainty for Trustworthy Human–AI Interaction

Why it matters

Hamed Hassani’s talk frames uncertainty as a way to govern how AI affects human decisions: systems should preserve sound human judgments while helping where people are likely to err. It also examines agentic AI, where support from humans and tools is treated as part of the…

Read more

Watch: https://www.youtube.com/watch?v=-EVhmnxoSGU

0

Understanding Language Model Hallucinations through Parametric Knowledge Integration

Why it matters

Yue Dong examines how language-model hallucinations can arise from failures in knowledge enrichment and answer extraction, then considers conflicts between a model’s parametric knowledge and information supplied in long contexts. The talk says s…

Read more

Watch: https://www.youtube.com/watch?v=dVFAEpvGvhM

0

Quantifying Reliability in AI: Hallucinations and Uncertainty

Why it matters

The talk frames quantifying reliability as central to trustworthy AI, then describes feature-based conformal techniques for detecting and quantifying hallucinations in image reconstruction and probabilistic regression with diffusion models for uncertainty quantification.…

Read more

Watch: https://www.youtube.com/watch?v=mabxnS40L1g

0

Understanding the Structure of Reasoning in Language Models

Why it matters

The talk explores how language models reason under uncertainty, allocate inference-time computation, and decide when to use external information or tools. It also examines why strong underlying capabilities may not translate into reliable reasoning behavior and how internal…

Read more

Watch: https://www.youtube.com/watch?v=gAwCSHujtGo

0

In Search of Missing Mass: Disentangling Discoveries from Hallucinations

Why it matters

Eric Nalisnick's talk examines when large language models can reliably contribute to scientific discovery. The description says he will discuss extending hallucination results to characterize discovery potential and how prompts likely to yield discoveries might…

Read more

Watch: https://www.youtube.com/watch?v=ZNlZsI9kBm4

0

Reliability Along the Way: Progress and Failure Signals in AI Agents

Why it matters

The talk examines how to detect whether LLM agents are progressing or failing during long-horizon tasks involving tools, irreversible actions, and unpredictable feedback. It presents the progress advantage from RL post-training as a step-level uncertainty signal,…

Read more

Watch: https://www.youtube.com/watch?v=5vBzKW3-wAQ

0

Why Pretrained Models Hallucinate

Why it matters

The talk analyzes language-model hallucinations mathematically and describes how next-word prediction training can induce them even when models are trained on clean data. Its description also outlines a reduction from a “fact or hallucination” classification task to generation or prompt com…

Read more

Watch: https://www.youtube.com/watch?v=gp8tmJHkE7c

0

Aligning Evaluation Incentives to Reduce Hallucination

Why it matters

The talk explains how accuracy-based LLM evaluation can reward confident guessing while failing to credit useful uncertainty. It discusses evaluation approaches that account for abstentions and the differing costs of mistakes, correct answers, and unanswered questions.

Discuss:…

Read more

Watch: https://www.youtube.com/watch?v=sGrnM2bOEHQ

0

LabOS: The AI-XR Co-Scientist That Sees and Works With Humans

Why it matters

The video presents LabOS as an AI-XR co-scientist designed to see and work with humans. Its description identifies the talk as addressing trustworthy AI, from hallucinations to reliable autonomy.

Discuss: What would an AI-XR co-scientist need to do to work reliably with…

Read more

Watch: https://www.youtube.com/watch?v=T_Cib6ACgGo

0

The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence, Task Complexity, and

Why it matters

Jascha Sohl-Dickstein of Anthropic presents a talk on trustworthy AI, framed around hallucinations and reliable autonomy. The title raises how misalignment may scale with model intelligence and task complexity.

Discuss: How should concerns…

Read more

Watch: https://www.youtube.com/watch?v=CsaNTn6NY08

0

Sequential Statistical Methods for Reliable Machine Learning

Why it matters

The talk presents sequential hypothesis tests for continuously auditing deployed machine-learning systems, with Type I error guarantees to detect performance degradation. It also describes parameter-free online conformal prediction for adaptive, group-conditional…

Read more

Watch: https://www.youtube.com/watch?v=j3Qumv16vk0

0

Grounding Diffusion Models: The Path to Reliable Synthesis Without Hallucination

Why it matters

The talk is titled “Grounding Diffusion Models: The Path to Reliable Synthesis Without Hallucination” and is presented under the broader theme of trustworthy AI, from hallucinations to reliable autonomy. The supplied metadata frames grounding as a pat…

Read more

Watch: https://www.youtube.com/watch?v=Cx47YnPmAeI

0

ATHENA: Steering Generative Models toward Better Visual Reasoning

Why it matters

The talk is titled “ATHENA: Steering Generative Models toward Better Visual Reasoning” and is framed within a series on trustworthy AI, from hallucinations to reliable autonomy. The supplied metadata does not specify the method or examples discussed.

Discuss: What app…

Read more

Watch: https://www.youtube.com/watch?v=soGaEIqJRnE

0

Uncertainty aware Causal Decision Making via Effect Bound Decomposition

Why it matters

Under the Trustworthy AI theme, this talk is titled around uncertainty-aware causal decision making via effect-bound decomposition. The description identifies Murat Kocaoglu of Johns Hopkins University as the speaker.

Discuss: How might effect-bound…

Read more

Watch: https://www.youtube.com/watch?v=-NfjZ_pyQRg

0

When Can We Trust What a Model Has Learned? Identifiability, Causality, and Reliable AI

Why it matters

The talk examines why models can perform well yet behave unreliably under distribution shifts or interventions, linking the problem to non-identifiability. It discusses how sparsity, environmental variation and causal assumptions may help recover…

Read more

Watch: https://www.youtube.com/watch?v=kYG4bd5SEgw

0