Study examines how decoding algorithms affect fairness in language generation models.
problem Impact of decoding algorithms on fairness in open-ended language generation.
method Systematic analysis of top-p, top-k, and temperature decoding algorithms. result Decoding algorithms significantly impact fairness across demographic groups.
CausalEvolve improves efficiency and discovery in open-ended scientific tasks.
problem Lack of targeted guidance and knowledge organization in evolve-based agents.
method Causal scratchpad that identifies and reasons about guiding factors for evolution.
result Effective improvement in evolutionary efficiency and better solutions.
New method for evaluating LLMs reduces bias in open-ended evaluations.
problem Bias in Elo-based ratings of LLMs due to data redundancies.
method Proposes evaluation as a 3-player game and introduces novel solution concepts.
result Novel method leads to more robust and intuitive ratings.
This paper presents the beginnings of an automatic statistician, focusing on regression problems. Our system explores an open-ended space of statistical models to discover a good explanation of a data set, and then produces a detailed report with figures and natural-language text. Our approach treats unknown regression…
Shai is a 10B model for asset management tasks, outperforming baselines.
problem Improving performance in asset management tasks.
method Continuous pre-training and fine-tuning on asset management-specific data.
result Shai outperforms baseline models in asset management tasks.
A new framework evaluates LLM calibration in open-ended QA.
problem Evaluating LLM calibration in open-ended QA settings.
method Sem-ECE framework: sampling answers, grouping by semantic classes, and using frequencies as confidence.
result Sem-ECE estimators are unbiased and Sem2 achieves smaller calibration error. Quriosity analyzes curiosity-driven questions from diverse sources.
problem Understanding and analyzing human curiosity-driven questions.
method Collection of 13.5K questions from various sources, development of an iterative prompt improvement framework.
result 42% of questions are causal, revealing unique linguistic properties.
Hyperfitting improves LLM generation quality by enhancing diversity, contrary to simple temperature scaling.
problem Improving open-ended generation quality of LLMs with minimal fine-tuning effort.
method Demonstrates that hyperfitting, a phenomenon where LLMs are fine-tuned to near-zero training loss, enhances generation quality and mitigates repetition.
result Hyperfitting is distinct from temperature scaling and involves a dynamic, context-dependent rank reordering mechanism in the final transformer block.
MiniHack simplifies creation of complex RL environments.
problem Limited availability of challenging RL benchmarks.
method Develops a sandbox framework for easy RL environment design.
result MiniHack enables rapid creation of diverse RL testbeds.
New method predicts nonfactuality in LLM responses using semantic isotropy.
problem Assessing trustworthiness of long-form LLM responses efficiently.
method Semantic isotropy of text embeddings on the unit sphere.
result Higher semantic isotropy correlates with lower factual consistency.
Adaptive contrastive search improves text generation quality and diversity.
problem Decoding high-quality text from language model outputs.
method Adaptive contrastive search incorporating uncertainty-guided degeneration penalty.
result Enhanced text generation quality and diversity across different models and datasets.
Framework synthesizes programs for simulating complex models and estimating parameters.
problem Parameter estimation for complex models requires manual encoding of fixed model structures.
method Combines LLMs for program synthesis with neural simulation-based inference.
result Identifies plausible model families from open-ended prompts with high accuracy.
Variance-Calibrated Modulation (VCM) addresses the likelihood trap in LLMs by reshaping the probability distribution before truncation.
problem LLMs fall into the likelihood trap, leading to repetitive degeneration and vocabulary dullness.
method VCM reshapes the probability distribution before truncation through Contextual Searchlight and Adaptive Self-Debiasing.
result VCM mitigates the likelihood trap across open-ended generation, factual QA, and mathematical reasoning.
Top-H decoding improves text generation by balancing creativity and coherence.
problem Balancing creativity and coherence in text generation.
method Entropy-constrained minimum divergence problem, NP-hard ECMM problem, top-H decoding.
result Top-H decoding outperforms min-p sampling by up to 25.63% on creative writing benchmarks.
A new framework evaluates LLMs by considering judge reliability.
problem Evaluating LLMs without ground truth labels can lead to biased results.
method Introduces judge-specific discrimination parameters and estimates model quality and judge reliability.
result Improves agreement with human preferences and produces calibrated uncertainty quantification.
The study introduces anytime learning schedules for large language models without fixed horizons.
problem Training large language models without knowing the total training horizon.
method Theoretical analysis and weight averaging to create anytime learning schedules.
result Theoretical and empirical evidence shows that weight averaging with simple step sizes can achieve comparable final loss to well-tuned cosine schedules.
New method samples from LLM posterior for coherent, useful responses.
problem Hallucinations in large language models.
method Posterior sampling for conditional generation, with calibration.
result Achieves statistical guarantees with higher downstream utility.
GSR optimizes tasks in scientific workflows, improving performance across diverse applications.
problem Uncertainty in task selection and evaluation in scientific workflow optimization.
method Generate-Select-Refine (GSR) framework that alternates between task generation and optimization.
result GSR outperforms existing LLM-based optimizers in various scientific applications.
Bayesian approach quantifies uncertainty in LLM evaluations.
problem Statistical uncertainty in evaluating LLM behavior.
method Bayesian evaluation of LLM behavior using probabilistic text generation strategies.
result Bayesian approach provides useful uncertainty quantification about LLM behavior.
We consider visual domains in which a class label specifies the content of an image, and class-irrelevant properties that differentiate instances constitute the style. We present a domain-independent method that permits the open-ended recombination of style of one image with the content of another. Open ended simply me…
The paper analyzes how open-end fund sales affect prices and returns.
problem How open-end fund sales impact prices and returns.
method Continuous-time market-clearing model to derive expected-return restrictions.
result Forced-sale pressure predicts actual fund selling and positive returns.
Meta-learning agents excel at rapidly learning new tasks from open-ended task distributions; yet, they forget what they learn about each task as soon as the next begins. When tasks reoccur - as they do in natural environments - metalearning agents must explore again instead of immediately exploiting previously discover…
A new approach reduces the size of prediction sets in CP, making it more efficient.
problem Large and costly prediction sets in standard CP.
method Expand correctness criterion to include inferred admissible answers, and use prediction cascades to prune implausible labels.
result Empirically effective in natural language processing and drug discovery.
Statistical framework improves LLM chatbot ranking.
problem Improving evaluation of LLM-based chatbots through pairwise comparisons.
method Factored tie model, covariance modeling, and parameter constraints.
result Substantial improvements in modeling pairwise comparison data.
Reward collapse occurs when ranking-based reward models yield uniform rewards for different prompts.
problem Reward collapse in aligning large language models with human preferences.
method Introduced a prompt-aware optimization scheme to derive closed-form expressions for reward distributions.
result Our prompt-aware utility functions significantly alleviate reward collapse during training.
Local-HDP learns independent topics for each 3D object category in real-time.
problem Learning independent topics for each 3D object category in real-time.
method Local-Hierarchical Dirichlet Process (Local-HDP) with online variational inference.
result Local-HDP outperforms other approaches in accuracy, scalability, and memory efficiency.
Robot learns multiple tasks hierarchically by transferring knowledge.
problem Learning multiple complex tasks in open-ended environments.
method Task-oriented procedures, goal-babbling, imitation learning, active learning, intrinsic motivation.
result Robots can learn complex tasks more efficiently by transferring knowledge from simpler ones.
Exposure bias has been regarded as a central problem for auto-regressive language models (LM). It claims that teacher forcing would cause the test-time generation to be incrementally distorted due to the training-generation discrepancy. Although a lot of algorithms have been proposed to avoid teacher forcing and theref…
FSPO optimizes synthetic preferences for LLM personalization.
problem Personalizing large language models for diverse users.
method FSPO reframes reward modeling as a meta-learning problem, using few labeled preferences and synthetic data.
result FSPO achieves high winrates in personalized responses, both synthetic and real.
A polynomial knot is a smooth embedding κ:ℜ→ℜn whose components are polynomials. The case n=3 is of particular interest. It is both an object of real algebraic geometry as well as being an open ended topological knot. This paper contains basic results for these knots as well as many examples.
There is a growing interest and literature on intrinsic motivations and open-ended learning in both cognitive robotics and machine learning on one side, and in psychology and neuroscience on the other. This paper aims to review some relevant contributions from the two literature threads and to draw links between them. …
New text-to-image diffusion models improve scene understanding for AI agents.
problem Fine-grained scene understanding for AI agents from text and images.
method Pre-trained text-to-image diffusion models optimized for generating images from text prompts.
result Policies learned with Stable Control Representations outperform state-of-the-art approaches on various control tasks.
New optimizer SF-NorMuon matches tuned AdamW across various horizons.
problem Fixed learning-rate schedules in neural network training lead to strong path dependence and costly re-tuning.
method Schedule-Free Spectral Optimization (SF-NorMuon)
result SF-NorMuon outperforms tuned AdamW on 125M and 772M parameter models across different horizons.
Zero-sum games such as chess and poker are, abstractly, functions that evaluate pairs of agents, for example labeling them `winner' and `loser'. If the game is approximately transitive, then self-play generates sequences of agents of increasing strength. However, nontransitive games, such as rock-paper-scissors, can ex…
DIVA generates diverse tasks for complex simulators, enabling adaptive agent training.
problem Lack of diverse training data for complex, open-ended simulators.
method Evolutionary approach using domain randomization and procedural generation.
result Successfully trains adaptive agent behavior in complex simulators.
AURA: Adaptive Uncertainty-aware Refinement for LLM-as-a-Judge Auditing
problem Auditing LLM-as-a-Judge decisions
method Adaptive uncertainty-aware refinement
result Human-consistent signal learning and evidence propagation
This paper is a survey on the theory of knotoids and braidoids. Knotoids are open ended knot diagrams in surfaces and braidoids are geometric objects analogous to classical braids, forming a counterpart theory to the theory of knotoids in the plane. We survey through the fundamental notions and existing works on these …
New homology categorifies knotoid polynomial.
problem Generalizing knot theory to open-ended diagrams.
method Defined winding homology as a triply-graded invariant.
result Categorifies Turaev's polynomial invariant of knotoids.
This work uses image generation models to find vision model bugs.
problem Automatically discovering failures in vision models.
method Conditional text-to-image generation and captioning models.
result Demonstrated utility of large-scale generative models to find vision model bugs.
This paper gives mathematical models for flat knotted ribbons, and makes specific conjectures for the least length of ribbon (for a given width) needed to tie the trefoil knot and the figure eight knot. The first conjecture states that (for width one) the least length of ribbon needed to tie an open-ended trefoil knot …
Hallucinations in models are mislinked estimates, not errors.
problem Hallucinations in generative models as failures to link estimates to plausible causes.
method Formalized hallucinations, showed even optimal estimators hallucinate, provided a general lower bound on hallucinate rate, reframed hallucination as structural misalignment, and experimentally supported theory.
result Hallucinations are structural misalignments between loss minimization and human-acceptable outputs, leading to estimation errors.
Graphoids are topological invariants of virtual graph diagrams.
problem Understanding knotted graphs with open ends in proteins and simplifying virtual spatial graphs.
method Topological interpretations of graphoids using graph Reidemeister moves.
result Virtual graphoids are useful for studying knotted graphs and simplifying spatial graphs.
New F-polynomial distinguishes knotoid diagrams not previously possible.
problem Polynomial invariants for knotoids.
method Introduced F-polynomial and constructed distinguishing examples. result New invariant F distinguishes knotoids not previously possible. Neural network-based Open-ended conversational agents automatically generate responses based on predictive models learned from a large number of pairs of utterances. The generated responses are typically acceptable as a sentence but are often dull, generic, and certainly devoid of any emotion. In this paper, we present…
Study finds Indian mutual funds adjust cash holdings based on inflows, impacting stock purchases.
problem Active liquidity management by mutual funds in India.
method Examined cash holdings and stock purchases of Indian equity mutual funds.
result Funds with active liquidity choices outperform, highlighting the importance of this strategy.
In massive open online courses (MOOCs), peer grading serves as a critical tool for scaling the grading of complex, open-ended assignments to courses with tens or hundreds of thousands of students. But despite promising initial trials, it does not always deliver accurate results compared to human experts. In this paper,…
New approach uses 'growth' and 'harvesting' concepts to improve deep learning models.
problem Current deep learning models lack transparency and high convergence rates.
method Reconsider neural networks as single-species population dynamics with balanced growth and harvesting rates.
result SGD with balanced growth and harvesting rates outperforms adaptive methods in all three requirements.
Autonomous multiple tasks learning is a fundamental capability to develop versatile artificial agents that can act in complex environments. In real-world scenarios, tasks may be interrelated (or "hierarchical") so that a robot has to first learn to achieve some of them to set the preconditions for learning other ones. …