Equal Ranking Quality, Different Decisions: Measuring and Reducing Order Dependence in LLM Scorers

LLM scorers with equal ranking quality make unstable threshold and preference decisions under candidate reordering, and order-consistency fine-tuning fixes it without harming quality.

Markus Frohmann, Mahdiyar Alavi, Elizabeth Lingg, Navid Rekabsaz

Published Aug 27, 2026 · 0 citations · ▲ 17 on Hugging Face · Code ★ 2

Attention Is All You Need

The Transformer replaces recurrence and convolutions with attention, achieving superior translation quality and faster training.

Ashish Vaswani, Noam Shazeer, Niki Jitendra Parmar, Jakob Uszkoreit and 4 more

Published Aug 23, 2025 · 26,828 citations

Laplacian Heads Improve Transformers by Smoothing Token Representations

Yuchong Zhang, Vardan Papyan

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Sequential Probability Assignment against Smoothed Adversaries with Unknown Base Measure

Ziyi Liu, Dan Roy

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Integrated Imputation-Classification for Supervised Learning with Missing Data

Yue Liu, Ben Liang, Ali Tizghadam, Ilijc Albanese

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

When to Trust a PFN: Detecting Harmful Shift in Tabular Foundation Models

Viet Nguyen, Herman Bergström, Stephan Rabanser, Rahul Krishnan

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Colour me shocked: Exact Molecular Hessians from MLIPs in O(N) time using sparse differentiation!

Luca Thiede, Andreas Burger, Alan Aspuru-Guzik

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

DynaSub: Adaptive Subgrouping for Scalable Representation Learning

Tina Behrouzi, Sana Tonekaboni, Rahul Krishnan, Anna Goldenberg

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

SuperSycophantic: Stress-Testing Frontier LLMs from Single- to Multi-Turn Sycophancy

Terry J Zhang, Oscar S Yasunaga, Wenyuan Jiang, Jessica Bo and 7 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

FedLoVA: Value-Only Aggregation for Federated LoRA Fine-Tuning of Large Language Models

Ensieh Khazaei, Baturalp Buyukates, Dimitrios Hatzinakos

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

On the Selectivity of Generative Models in Structure-Based Drug Design

Ella Miray Rajaonson, Jungyoon Lee, William Chau, Alan Aspuru-Guzik and 4 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

STAR-Math: Multi-Agent Mathematical Reasoning under Persistent Meta-Strategic Supervision

Jiaao Wu, Xian Zhang, Hanzhang Liu, Sophia Zhang and 2 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Plan2Sense: Open-World Task Planning in Epistemic States via Interleaved Ontic and Sensing Actions

Xiaotian Liu, Armin Toroghi, Jiazhou Liang, Ali Pesaranghader and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Runtime Verification of Multiple Natural Language Criteria for Agent Governance

Silviu Pitis, Parand A. Alamdari, Jessica Tang, Toryn Klassen and 1 more

Atlanta Poster Session 6, Fri, Dec 11, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Towards Self-Supervised, Generalizable and Decomposable 4D Driving Scene Reconstruction

James Tu, Anqi Joyce Yang, Jingkang Wang, Sivabalan Manivasagam and 1 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

What should post-training optimize? A test-time scaling law perspective

Muheng Li, Jian Qian, Wenlong Mou

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Hit Expansion via Localized Exploration of Synthesizable Chemical Space

Walter Virany, Yidong Jin, Andrew Lian, Dmytro Shevchuk and 5 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Think about how AIs think about themselves

Raymond Douglas, Jan Kulveit, Ondřej Havlíček, Theia Pearson-Vogel and 1 more

Paris Poster Session 5, Fri, Dec 11, 11:30 AM–1:30 PM, Paris Poster Hall · Published 2026

Median-of-Means under Structured Heavy-Tailed Noise: High-Probability Bounds for Clipped Stochastic Optimization

Ahmed El Bajdali, Ohad Shamir, Samuel Horváth, Eduard Gorbunov

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

The Heel of RLVR: Benchmark Glory Should Not Outpace Honest Measurement

Shuo Yang, Chiyu Ma, Kexin Huang, Jinda Lu and 10 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

VerifyThisBench: Joint Evaluation of Code, Specifications, and Proof

Xun Deng, Barış Bayazıt, Si Cheng Zhong, Andreas Veneris and 2 more

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Precise Debugging Benchmark: Is Your Model Debugging or Regenerating?

Miaosen Chai, Wang Bill Zhu, Shangshang Wang, Yejia Liu and 4 more

Atlanta Poster Session 2, Wed, Dec 9, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Ground False: Uncovering Errors in Formal Mathematics Benchmarks

Marcus Min, One An, Xujie Si, Osbert Bastani

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

Literati: Towards Anytime Optimal Shape Generalized Trees via AO*

Nakul Upadhya, Eldan Cohen

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Reinforcement Learning Agents Are Swimmers

Juan Rojas, Jacob Adamczyk, Abhishek Naik, Volodymyr Makarenko and 5 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Foundation Model Informed Acquisition Functions for Molecular Discovery

Qi CHEN, Fabio Ramos, Alan Aspuru-Guzik, Florian Shkurti

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

UTOPI: Efficient Egocentric Long-Video Understanding in AR via User-Guided Token Pre-Compression

ziqi wang, Su Chen, Qiance Tang, Jieyu Lin and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Test-time Scaling of Diffusions with Flow Maps

Flow Map Trajectory Tilting uses flow maps to enable principled diffusion test-time scaling with reward gradients, improving reward ascent and enabling complex image editing via vision-language models.

Amirmojtaba Sabour, Michael Albergo, Carles Domingo i Enrich, Nicholas Boffi and 3 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 6 on Hugging Face

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

DD-Ranking reveals dataset distillation gains come from extra evaluation techniques rather than image quality, proposing fair metrics to assess true synthetic dataset value.

Zekai Li, Xinhao Zhong, Samir Khaki, Zhiyuan Liang and 36 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

Multi-Marginal Couplings for Metropolis--Hastings

Multi-marginal coupling of Metropolis-Hastings chains via shared-randomness Poisson Monte Carlo improves coalescence rates and reduces meeting times up to 50%.

Truong Buu Phan, Gergely Flamich, Ashish Khisti, Shahab Asoodeh

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

TurboVGGT: Fast Visual Geometry Reconstruction with Adaptive Alternating Attention

TurboVGGT enables fast multi-view 3D reconstruction via adaptive alternating attention that balances sparse global and local frame attention while maintaining competitive quality.

David Huang, Guile Wu, Chengjie Huang, Bingbing Liu and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

MedKIT: Evaluating Knowledge Integration and Generalization in Large Language Models

MedKIT evaluates medical LLM knowledge integration via clinical updates, revealing strong recall but limited relational, compositional, and operational generalization across 12 strategies.

Lukas Thede, Yash Kumar, David Chen, Danielle Bitterman and 3 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

DriveDreamer-Policy: A Geometry-Grounded World-Action Model for Unified Generation and Planning

DriveDreamer-Policy unifies depth generation, video prediction, and motion planning via geometry-aware world representations, achieving 89.2 PDMS on Navsim v1 and 88.7 EPDMS on v2.

Yang Zhou, Xiaofeng Wang, Hao Shao, Letian Wang and 7 more

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 2 on Hugging Face · Code ★ 60

AVIS: Adaptive Test-Time Scaling for Vision–Language Models

AVIS introduces a per-query adaptive policy that jointly scales visual token pruning and reasoning rollouts to improve vision-language model accuracy-compute trade-offs.

Ahmadreza Jeddi, Minh Le, Amirhossein Kazerouni, Hakki Karaimer and 7 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Addressable Memory for Video World Models

Interactive video world models lose visual persistence beyond training horizons because temporal RoPE offsets become out-of-distribution; WorldTrace assigns compressed memory slots virtual in-distribution positions to restore addressability, boosting temporal consistency by 15.5% and episodic recall

Xindi Wu, Sven Elflein, James Lucas, Olga Russakovsky and 4 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 16 on Hugging Face

P$^{3}$: Joint Program-and-Proof Planning\\ for Verified Code Generation

P³ plans programs and proofs jointly from specifications before elaboration, outperforming sequential baselines by up to 11.2 points on verified generation benchmarks while reducing cost and time.

Zenan Li, Ziran Yang, Peiyang Song, Zhaoyu Li and 1 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

SGD in Multiclass Logistic Regression: Sequential Learning and Scaling Laws

Multiclass logistic regression learns Gaussian mixtures sequentially by class frequency, producing power-law risk phases and compute-optimal scaling laws.

Konstantinos Tsiolis, Denny Wu, Christos Thrampoulidis, Murat Erdogdu

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Efficient Prediction of Pass@k Scaling in Large Language Models

Standard pass@k scaling laws suffer statistical shortcomings, so a beta-binomial framework and dynamic sampling strategy more accurately predict rare LLM capabilities and risks from limited data.

Joshua Kazdan, Colin Sullivan, Rylan Schaeffer, Youssef Allouah and 3 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Negation Neglect: When models fail to learn negations in training

Fine-tuning LLMs on documents that flag claims as false makes them believe those claims, with belief rates jumping from 2.5% to 88.6%, though local negation phrasing largely prevents it.

Harry Mayne, Lev McKinney, Jan Dubiński, Adam Karvonen and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Vision Transformers Learn Gestalt-Like Figure-Ground Cues from Natural Images

Vision transformers learn Gestalt-like figure-ground cues, surroundedness, convexity, and symmetry for uniform regions, from natural images, with linear probes generalizing zero-shot to artificial stimuli.

Matthias Tangemann, Benjamin Lo, Zygmunt Pizlo, Kaleem Siddiqi and 2 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

X-Palm: Paired Multispectral-to-Smartphone Dataset for Cross-Domain Palmprint Authentication

X-Palm is a cross-domain palmprint dataset pairing controlled multispectral and unconstrained smartphone images that reveals severe performance collapse of existing models on real-world mobile authentication.

Seyed Jamal Seyedmohammadi, Pai Chet Ng, Angelo Genovese, Zhixiang Chi and 2 more

Paris Poster Session 2, Wed, Dec 9, 5:00 PM–7:00 PM, Paris Poster Hall · Published 2026

Wasserstein residuals: Learning Gradient Flows from Population Dynamics

A residual-based continuity loss for Wasserstein gradient flows yields a simulation-free stitching method robust to sparse observations and state-of-the-art on trajectory inference benchmarks.

Markus Heinonen, Yair Shenfeld, Ricardo Baptista, Daniel Waxman and 3 more

Atlanta Poster Session 3, Thu, Dec 10, 10:00 AM–1:00 PM, Hall C1 · Published 2026

PiD: Fast and High-Resolution Latent Decoding with Pixel Diffusion

PiD reformulates latent decoding as conditional pixel diffusion to synthesize high-resolution images with low latency and high fidelity. It decodes 512×512 latents to 2048×2048 pixels in under one second on consumer GPUs.

Yifan Lu, Qi Wu, Jay Zhangjie Wu, Zian Wang and 3 more

Sydney Poster Session 3, Wed, Dec 9, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 45 on Hugging Face

Predicting and improving test-time scaling laws via reward tail-guided search

Tail-guided reward estimation predicts LLM test-time scaling laws and guides SLG Search to dynamically allocate compute, achieving vanishing regret and polynomial compute savings over best-of-N.

Muheng Li, Jian Qian, Wenlong Mou

Atlanta Poster Session 4, Thu, Dec 10, 4:30 PM–7:30 PM, Hall C1 · Published 2026

Adaptively Incorporating Directional Hints into Zeroth-Order Optimization

CV-ZOD adaptively integrates directional hints into zeroth-order optimization, achieving rates that interpolate between first- and zeroth-order convergence based on hint quality without prior knowledge.

Alexander Ryabchenko, Jian Qian, Wenlong Mou

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Persona Generators: Generating Diverse Synthetic Personas for Arbitrary Contexts

Persona Generators use evolutionary code optimization to expand brief context descriptions into diverse synthetic populations maximizing opinion and preference coverage. Evolved generators substantially outperform baselines across six diversity metrics by spanning rare trait combinations.

Davide Paglieri, Logan Cross, William Cunningham, Joel Leibo and 1 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Fast Organic Crystal Structure Prediction with Unit Cell Flow Matching

Clari predicts organic crystal structures via unit-cell flow matching with pure pair-bias attention, cutting generation to seconds while surpassing OXtal solve rates and supporting non-sanitizable inputs.

Alston Lo, Luka Mucko, Austin Cheng, Andy Cai and 3 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 2 on Hugging Face

SEGA: Spectral-Energy Guided Attention for Resolution Extrapolation in Diffusion Transformers

SEGA adaptively scales diffusion-transformer attention by latent frequency to improve high-resolution synthesis without training.

Javad Rajabi, Kimia Shaban, Koorosh Roohi, David Lindell and 1 more

Paris Poster Session 2, Wed, Dec 9, 5:00 PM–7:00 PM, Paris Poster Hall · Published 2026 · ▲ 43 on Hugging Face · Code ★ 74

GitInject: Real-World Prompt Injection Attacks in AI-Powered CI/CD Pipelines

GitInject tests real AI CI/CD workflows and finds all providers vulnerable to prompt injection via structural credential and config handling flaws.

Jafar Isbarov, Umid Suleymanov, I Shumailov, Murat Kantarcioglu

Paris Poster Session 6, Fri, Dec 11, 2:30 PM–4:30 PM, Paris Poster Hall · Published 2026

Blocked Gibbs meets Diffusion Transformers: Unsupervised Learning for Constraint Optimization

BloGDiT replaces joint Gaussian denoising with blocked Gaussian diffusion using iterative block resampling and annealed block sizes to enable targeted edits for constraint optimization, matching or outperforming prior methods on Sudoku, graph coloring, MIS, and MaxCut.

Yudong Will Xu, Wenhao Li, Xiaoyu Wang, Scott Sanner and 1 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

ProxyPose: 6-DoF Pose Tracking via Video-to-Video Translation

ProxyPose recasts 6-DoF pose tracking as video-to-video translation using a diffusion model to generate proxy videos for classical pose estimation, achieving state-of-the-art accuracy without 3D models or masks.

Ruihang Zhang, Felix Taubner, Pooja Ravi, Kyros Kutulakos and 1 more

Sydney Poster Session 5, Thu, Dec 10, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026

mRNABench: A curated benchmark for mature mRNA property and function prediction

mRNABench benchmarks mature mRNA property predictions across 59 tasks and 135K experiments, revealing synergies between self-supervised objectives that yield a compact state-of-the-art Mamba model using 700x fewer parameters.

Ruian (Ian) Shi, Taykhoom Dalal, Philip Fradkin, Divya Koyyalagunta and 9 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

JMed48k: A Multi-Profession Japanese Medical Licensing Benchmark for Vision-Language Model Evaluation

JMed48k introduces a Japanese medical licensing benchmark with 48,862 questions showing proprietary vision-language models gain substantially from images while medical-specific systems ignore visual evidence.

Yue Xun, Junyu Liu, Qian Niu, Xinyi Wang and 9 more

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

A Measure-Theoretic Analysis of Reasoning: Structural Generalization and Approximation Limits

Reasoning is formalized via optimal transport to bound transformers' OOD generalization by architectural Lipschitz continuity and approximation limits, proving depth is needed for backtracking and shift-invariant attention reduces risk.

Yuyang Zhang, Yifu Zhang, Xuehai Zhou, Xiaoyin Chen

Sydney Poster Session 6, Thu, Dec 10, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

PATCH: Learnable Tile-Level Hybrid Sparsity for LLMs

PATCH learns tile-level hybrid sparsity mixing dense and 2:4 tiles for LLMs, enabling tunable sparsity ratios that improve accuracy and deliver 1.18x-1.38x speedups over dense models.

Mohammad Mozaffari, Younes Hourri, Maryam Mehri Dehnavi

Sydney Poster Session 1, Tue, Dec 8, 10:00 AM–1:00 PM, Hall 1-4 · Published 2026 · ▲ 1 on Hugging Face

C3P: Contrastive promoter-protein pretraining yields representations capturing bacterial gene regulation

C3P uses contrastive promoter-protein pretraining to learn bacterial promoter representations that outperform genome language models on regulatory inference and zero-shot co-regulated gene retrieval.

Cameron Dufault, Scott Xu, Alan Moses

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026

Squeezing Capacity from Multimodal Large Language Models for Subject-driven Generation

Conditioning diffusion models on multimodal large language models with VAE identity conditioning and dual-layer aggregation improves subject-driven generation by balancing semantics with identity preservation.

Shuhong Zheng, Aashish K Misraa, Kevin Li, Yu-Jhe Li and 1 more

Sydney Poster Session 2, Tue, Dec 8, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026 · ▲ 7 on Hugging Face

OpenMedReason: Scientific Reasoning Supervision for Medical Vision–Language Models

OpenMedReason is a 450K-instance open medical reasoning dataset derived from scientific articles that improves LVLM diagnostic accuracy by 20% and enhances perception, knowledge, and reasoning.

Negin Baghbanzadeh, Pritam Sarkar, Michael Colacci, Abeer Badawi and 5 more

Paris Poster Session 4, Thu, Dec 10, 5:30 PM–7:30 PM, Paris Poster Hall · Published 2026

Mini Amusement Parks (MAPs): A Testbed for Modelling Business Decisions

MAPs introduces a mini amusement-park simulator benchmarking integrated business decision-making, finding experts outperform state-of-the-art agents by over 11x due to weaknesses in long-horizon planning, sample-efficient learning, and spatial reasoning.

Stéphane Aroca-Ouellette, Ian Berlot-Attwell, Panagiotis Lymperopoulos, Abhiramon Rajasekharan and 4 more

Atlanta Poster Session 1, Wed, Dec 9, 10:00 AM–1:00 PM, Hall C1 · Published 2026

Neural Proposals, Symbolic Guarantees: Neuro-Symbolic Graph Generative Modeling

A neuro-symbolic framework pairs neural graph proposals with symbolic SMT solvers for hard-constraint satisfaction, achieving over 95% in-distribution and 64, 86% zero-shot rule compliance on the MolSAT benchmark.

Chuqin Geng, Li Zhang, Mark Zhang, Zhaoyue(Rebecca) Wang and 2 more

Sydney Poster Session 4, Wed, Dec 9, 5:00 PM–8:00 PM, Hall 1-4 · Published 2026