Goto

Collaborating Authors

 Industry


RoSHAP: A Distributional Framework and Robust Metric for Stable Feature Attribution

arXiv.org Machine Learning

Feature attribution analysis is critical for interpreting machine learning models and supporting reliable data-driven decisions. However, feature attribution measures often exhibit stochastic variation: different train--test splits, random seeds, or model-fitting procedures can produce substantially different attribution values and feature rankings. This paper proposes a framework for incorporating stochastic nature of feature attribution and a robust attribution metric, RoSHAP, for stable feature ranking based on the SHAP metric. The proposed framework models the distribution of feature attribution scores and estimates it through bootstrap resampling and kernel density estimation. We show that, under mild regularity conditions, the aggregated feature attribution score is asymptotically Gaussian, which greatly reduces the computational cost of distribution estimation. The RoSHAP summarizes the distribution of SHAP into a robust feature-ranking criterion that simultaneously rewards features that are active, strong, and stable. Through simulations and real-data experiments, the proposed framework and RoSHAP outperform standard single-run attribution measures in identifying signal features. In addition, models built using RoSHAP-selected features achieve predictive performance comparable to full-feature models while using substantially fewer predictors. The proposed RoSHAP approach improves the stability and interpretability of machine learning models, enabling reliable and consistent insights for analysis.


Text Knows What, Tables Know When: Clinical Timeline Reconstruction via Retrieval-Augmented Multimodal Alignment

arXiv.org Machine Learning

Reconstructing precise clinical timelines is essential for modeling patient trajectories and forecasting risk in complex, heterogeneous conditions like sepsis. While unstructured clinical narratives offer semantically rich and contextually complete descriptions of a patient's course, they often lack temporal precision and contain ambiguous event timing. Conversely, structured electronic health record (EHR) data provides precise temporal anchors but misses a substantial portion of clinically meaningful events. We introduce a retrieval-augmented multimodal alignment framework that bridges this gap to improve the temporal precision of absolute clinical timelines extracted from text. Our approach formulates timeline reconstruction as a graph-based multistep process: it first extracts central anchor events from narratives to build an initial temporal scaffold, places non-central events relative to this backbone, and then calibrates the timeline using retrieved structured EHR rows as external temporal evidence. Evaluated using instruction-tuned large language models on the i2m4 benchmark spanning MIMIC-III and MIMIC-IV, our multimodal pipeline consistently improves absolute timestamp accuracy (AULTC) and improves temporal concordance across nearly all evaluated models over unimodal text-only reconstruction, without compromising event match rates. Furthermore, our empirical gap analysis reveals that 34.8% of text-derived events are entirely absent from tabular records, demonstrating that aligning these modalities can produce a more temporally faithful and clinically informative reconstruction of patient trajectories than either source alone.


UN aid convoy hit by drone strikes in Ukraine's Kherson

Al Jazeera

What are Russia's gains from the Iran war? 'We are not losers; we are winners' UN aid convoy hit by drone strikes in Ukraine's Kherson NewsFeed UN aid convoy hit by drone strikes in Ukraine's Kherson A UN humanitarian convoy delivering aid to the city of Kherson was hit twice by drones, despite prior coordination with Ukrainian and Russian forces. No injuries were reported, and the UN has not attributed the attack to either side. 'China is gaining from what the US is doing in Iran' Iran's FM urges BRICS states to condemn US-Israeli aggression


High-stakes courtroom drama of Musk v OpenAI hears closing arguments

The Guardian

OpenAI's CEO, Sam Altman, arrives at the federal courthouse in Oakland, California, on Thursday. OpenAI's CEO, Sam Altman, arrives at the federal courthouse in Oakland, California, on Thursday. Nine-person jury to consider whether AI firm bilked world's richest person and unjustly enriched themselves Closing arguments began on Thursday in Elon Musk's lawsuit against Sam Altman and OpenAI, bringing the weeks-long courtroom battle between the two tech moguls nearer to a decision. A nine-person jury is set to deliberate and return a verdict on whether they believe the AI firm and Altman are liable in the case. The trial, which began last month in an Oakland, California, federal courthouse, has gripped Silicon Valley and featured some of the tech industry's biggest names as witnesses.


AI could put people off tech jobs and hurt the economy, warns Raspberry Pi boss

BBC News

The founder of British computer maker Raspberry Pi has warned that overestimating the abilities of Artificial Intelligence (AI) could put people off pursuing tech jobs and hurt the economy. Eben Upton told the BBC's Big Boss Interview podcast this could distort people's choices in ways that make that skill shortage worse and not better. Some people are very inclined to overestimate what these [AI] tools can do, he said, and warned against claims that it would destroy vast numbers of computing roles over the coming years. The rise of tools such as ChatGPT and Claude have led to predictions of huge job losses, particularly for tech workers and graduates. Amazon, Meta and Microsoft have already blamed tens of thousands of layoffs on AI over the last year.


While Donald Trump Adventures in China, D.C. Entertains Itself

The New Yorker

The President swept off to Beijing to court Xi Jinping. Back Stateside, it was non-Presidential motorcades, video games, and a languid vibe at the White House. The endless motorcades and wail of sirens in Washington this week made it seem as if the President were travelling non-stop around the city, or receiving a bevy of foreign dignitaries. As it happened, it was National Police Week, and the ceremonial convoys were carrying the families of police officers who had been killed in the line of duty. Donald Trump was leaving for China.


Closing arguments begin in Elon Musk's landmark lawsuit against OpenAI

Al Jazeera

Closing arguments begin in Elon Musk's landmark lawsuit against OpenAI Lawyers for OpenAI and Elon Musk began closing arguments in a landmark trial that could impact the future of the ChatGPT maker. On Thursday, each side presented a concluding statement to jurors, who will decide whether OpenAI and its leaders profited from a venture that was meant to be a "charity". Musk sued OpenAI, its CEO Sam Altman and its president Greg Brockman, alleging that the company strayed from its founding mission to build AI that was safe and beneficial to humanity. Musk was not present for the closing statements on Thursday, as he is currently in China on a diplomatic visit with United States President Donald Trump. His lawyer, Steven Molo, used his final remarks to accuse OpenAI of breaching its charitable trust by enriching investors and insiders at the nonprofit's expense.


Japanese bear-fightin' robo-wolves are pure unleaded nightmare fuel but they're working

FOX News

Jennifer Lopez's dress holds on for dear life at her Netflix event, Trump powers through some wine & Kay Adams Eric Church's uses his guitar to deliver one of the most powerful addresses ever at UNC commencement Ella Langley crushes online troll with just four words, reminds the internet she doesn't miss Victoria's Secret should sign Rachel Pizzolato to face Sydney Sweeney in lingerie war, Reds fan is dumb & MEAT Morgan Wallen fan goes viral kicking a phone out of someone's hand as she's being escorted out in handcuffs Are teams that insist on singing'Sweet Caroline' during games the worst thing in sports? If this hasn't been said before, it should've been -- you can't hide in the bushes at a bachelorette pool party Shakira cranks up the heat with a World Cup song that has people dancing, buy Elvis' rhinestone jock & BBQ Greg Gutfeld: Maybe Katie Couric is coming around, who knows! Gutfeld: We don't have this kind of person on the right because we have humility Trump pledges to raise Jimmy Lai's case with Xi amid Hong Kong publisher's 20-year prison sentence Jesse Watters: Once the US gets fast food into a country, that's it Brit Hume: China's attitude on Taiwan has'never changed' History'judges' Obama for Iran nuclear deal, Iranian-American activist says Lt Gen Keith Kellogg says Iran is led by a'bunch of thugs' Larry Kudlow: 'The chickens are coming home to roost' OutKick Japanese bear-fightin' robo-wolves are pure unleaded nightmare fuel but they're working A humanoid robot chasing a pack of wild boars yeah, this actually happened. There are so many things happening around the world, it's really tough to stay up to speed on everything. However, one thing that I can't believe isn't talked about more is the problem Japan is having with bear attacks . In fact, it has gotten so bad that they've turned to robotic wolves to deal with it.


Sam Altman Is Taking a Lot of Punches on the Witness Stand

Mother Jones

Elon Musk's team seems to have one main goal: make the OpenAI boss look like a liar. Musk's wins so far mainly involve making OpenAI and Altman look ridiculous. Get your news from a source that's not owned and controlled by oligarchs. Can you trust Sam Altman? That was one of the central themes at the high-profile trial between the OpenAI CEO and Elon Musk in California this week, as Musk's lawyers peppered Altman with questions on his work relationships, including his temporary ouster from OpenAI three years ago by a mistrustful board of directors .


New DNA analysis of Christopher Columbus reveals truth about explorer's origins that rewrites history

Daily Mail - Science & tech

Marco Rubio warns China of'repercussions' as he reveals what really happened during closed-door Trump and Xi meeting Ex-Yankees star Carl Pavano'peed in shampoo bottles and soiled the bed,' ex-wife claims as bitter prenup feud takes disgusting twist Fury as Kash Patel SNORKELS at sacred war tomb where 900 sailors still lie... then jets off to Las Vegas Glamorous Texas Democrat's secret KINK exposed: Congressional candidate's past life returns to haunt her After theater groping shame, Lauren Boebert is being bankrolled by America's cringiest ex-congressman... and it exposes a MASSIVE hypocrisy Horrifying final days of killer dad Chris Watts' pregnant wife before she was slaughtered alongside their daughters. Read all the chilling texts and receipts in full for first time: 'My eyes burn from crying' RHOBH star Diana Jenkins denies claims she put Hayden Panettiere in bed with'undressed man' when she was 18 Trump reveals Xi's offer to break Iran's Hormuz chokehold... as China's price for the rescue looms Mystery blonde Trump aide with unfettered access to President's phone sparks White House friction: Real reason his posts contain random capital letters... and shadowy team behind them unmasked Despicable crimes paid for couple's lavish lifestyle that they flaunted online while gold chain-wearing husband fleeced $1BILLION from taxpayers New DNA analysis of Christopher Columbus reveals truth about explorer's origins that rewrites history Bitter cat fight erupts over DHS'sugar baby' scandal: Veteran female intelligence officer launches explosive new accusations that go right to top of counterterror HQ I lost 9lb in two weeks by making one simple tweak to my lifestyle. I didn't use Mounjaro, diet or change how I exercise and I couldn't believe the results... anyone can do it too I'm godfather to Candace Owens' daughter and Charlie Kirk was my friend... so I know the real reason she's attacking Erika - and I'll never publicly condemn her Britney Spears seen'barking and carrying knife' during chaotic restaurant visit I've had acid reflux all my life. Target customers threaten to boycott store after controversial'upgrade' to shopping cart New DNA analysis of Christopher Columbus reveals truth about explorer's origins that rewrites history A new DNA analysis of remains belonging to several direct descendants of Christopher Columbus may have uncovered a history-changing truth about the explorer's origins. For centuries, historians have believed the explorer was born in Genoa, Italy, rising from humble beginnings to persuade the Catholic Monarchs to finance what many considered an impossible voyage across the Atlantic.