Post-training, which elicits a pretrained Base model into the corresponding Instruct model, is widely considered to be superficial. In this work, we first reinforce this hypothesis by providing novel quantitative evidence from the weight level that the effective rank (eRank) remains negligibly changed. However, this superficiality also suffers a critical trade-off, improving the exploitation capabilities at the cost of limiting its exploration. To tackle this issue, we propose Timber, a simple yet effective training-free method that enhances the exploration capability of the Instruct model while preserving its exploitation. The key insight is to partially revert Instruct towards the paired Base model by subtle yet targeted refinement of the weight deltas. Extensive experiments on Llama and Qwen series demonstrate that Timber consistently improves vanilla Instruct models, particularly on Pass@k performance. Our findings offer new insights into the post-training stage at the weight level and practical strategies to refine the Instruct model without training. Large Language Models (LLMs), such as Qwen3 (Y ang et al., 2025), Llama 3 (Grattafiori et al., 2024), and Deepseek R1 (Guo et al., 2025), have achieved superior success in Natural Language Process (NLP), especially in reasoning tasks (Huang & Chang, 2022). To train these LLMs, a Base model is first pretrained on huge amounts of data. After that, a post-training stage is applied to train an Instruct model, adapting supervised finetuning (SFT) and reinforcement learning (RL) to elicit alignment and reasoning ability (Y ang et al., 2025). The post-training stage tends to be superficial, i.e., post-training only utilizes the pattern contained in the Base model acquired during pre-training (Y ue et al., 2025; Zhou et al., 2023a; Y e et al., 2025; Muennighoff et al., 2025). In this paper, we investigate the Base and Instruct models through the lens of effective rank (eRank, (Roy & V etterli, 2007)), providing a novel weight-level perspective on the superficiality of post-training. As shown in Figure 1, the eRanks of corresponding linear layers from the Base and Instruct models are almost identical. We can find that post-training induces only negligible changes to the effective dimensionality, offering new supporting evidence from the weight level for its superficiality.

We present Timber, the first white-box poisoning attack targeting decision trees. Timber is based on a greedy attack strategy leveraging sub-tree retraining to efficiently estimate the damage performed by poisoning a given training instance. The attack relies on a tree annotation procedure which enables sorting training instances so that they are processed in increasing order of computational cost of sub-tree retraining. This sorting yields a variant of Timber supporting an early stopping criterion designed to make poisoning attacks more efficient and feasible on larger datasets. We also discuss an extension of Timber to traditional random forest models, which is useful because decision trees are normally combined into ensembles to improve their predictive power. Our experimental evaluation on public datasets shows that our attacks outperform existing baselines in terms of effectiveness, efficiency or both. Moreover, we show that two representative defenses can mitigate the effect of our attacks, but fail at effectively thwarting them.

Significant advancements in the field of wood species identification are needed worldwide to support sustainable timber trade. In this work we contribute to automate the identification of wood species via high-resolution macroscopic images of timber. The main challenge of this problem is that fine-grained patterns in timber are crucial in order to accurately identify wood species, and these patterns are not properly learned by traditional convolutional neural networks (CNNs) trained on low/medium resolution images. We propose a Timber Deep Learning Identification with Patch-based Inference Voting methodology, abbreviated TDLI-PIV methodology. Our proposal exploits the concept of patching and the availability of high-resolution macroscopic images of timber in order to overcome the inherent challenges that CNNs face in timber identification. The TDLI-PIV methodology is able to capture fine-grained patterns in timber and, moreover, boosts robustness and prediction accuracy via a collaborative voting inference process. In this work we also introduce a new data set of marcroscopic images of timber, called GOIMAI-Phase-I, which has been obtained using optical magnification in order to capture fine-grained details, which contrasts to the other datasets that are publicly available. More concretely, images in GOIMAI-Phase-I are taken with a smartphone with a 24x magnifying lens attached to the camera. Our data set contains 2120 images of timber and covers 37 legally protected wood species. Our experiments have assessed the performance of the TDLI-PIV methodology, involving the comparison with other methodologies available in the literature, exploration of data augmentation methods and the effect that the dataset size has on the accuracy of TDLI-PIV.

In product design, a decomposition of the overall product function into a set of smaller, interacting functions is usually considered a crucial first step for any computer-supported design tool. Here, we propose a new approach for the decomposition of functions especially suited for later solutions based on Artificial Intelligence. The presented approach defines the decomposition problem in terms of a planning problem--a well established field in Artificial Intelligence. For the planning problem, logic-based solvers can be used to find solutions that compute a useful function structure for the design process. Well-known function libraries from engineering are used as atomic planning steps. The algorithms are evaluated using two different application examples to ensure the transferability of a general function decomposition.

High-intensity storms cause billions of pounds of damage every year, and climate change is set to make this worse in future. We already appear to be seeing more frequent and intense windstorms. Ex-hurricane Ophelia and Storm Eleanor both wreaked havoc in the British Isles over the winter, including injuries, power cuts and severe travel delays. It's not only commuters and households that are affected. Every year across Europe, the number of trees that commercial forests lose to storms is equivalent to the annual amount of timber felled in Poland.

High-intensity storms cause billions of pounds of damage every year, and climate change is set to make this worse in future. We already appear to be seeing more frequent and intense windstorms. Hurricane Ophelia and Storm Eleanor both wreaked havoc in the British Isles over the winter, including injuries, power cuts and severe travel delays. It's not only commuters and households that are affected. Every year across Europe, the number of trees that commercial forests lose to storms is equivalent to the annual amount of timber felled in Poland.

A home once built by Texas Gov. Greg Abbott is seen in Austin, Texas, Thursday, Aug. 10, 2017. While serving as state attorney general in 2011, Abbott tore down his Austin home and built the new one. City records show Abbott was allowed to do so as long as he didn't damage the root systems of two large pecan trees, though roots were eventually damaged in the renovations. As governor, Abbott has called tree ordinances like Austin's "socialistic."

Five actors gathered in a room on Lafayette Street, in downtown Manhattan, to start rehearsing a new work for the Public Theatre, "Joan of Arc: Into the Fire." Written by David Byrne, formerly of the Talking Heads, the show recast the enduring, improbable story of Joan--a teen-age girl in medieval France who experienced divine visions, led an army to defeat an occupying power, and was burned at the stake for heresy--as a rock musical that spoke to the current political moment. It was early January, and, that morning, U.S. intelligence officials had arrived at Trump Tower to brief the President-elect, Donald Trump, on the findings of an investigation into the recent election, in which they had concluded that President Vladimir Putin, of Russia, had acted to insure the defeat of Hillary Clinton. Inauguration Day was looming, and the rehearsal room had a troubled mood that reflected more than the ordinary anxieties of creating a show. The actors arranged four tables into a rectangle and sat down with Alex Timbers, the director of "Joan of Arc." Timbers, who is thirty-eight, is tall and fine-featured. He wore a denim shirt and black jeans that hung off his lanky, slightly hunched frame. His hair is dark and thick, and he frequently runs a hand through it, like a Romantic poet on deadline. Despite the air of disquiet, Timbers, who talks like a cool high-school teacher--lots of vocal fry, the repeated use of "awesome"--addressed the cast with rousing enthusiasm. He explained that, though the show had been in development for two years, it remained a work in progress. "I don't think anything is sacred--we are going to be building this together," Timbers said to the actors, all of whom were men except for Jo Lampert, a thirty-one-year-old newcomer, who was to play Joan. Timbers presented a scale model of the stage design, which had been conceived by Chris Barreca. When built, the set would be black and austere, and filled with enormous L.E.D. screens. A staircase extended from wing to wing, and at center stage there was a vertiginous platform. The set was on a turntable, and as it revolved it represented everything from a cathedral to a prison tower.

Although the use of metric fluents is fundamental to many practical planning problems, the study of heuristics to support fully automated planners working with these fluents remains relatively unexplored. The most widely used heuristic is the relaxation of metric fluents into interval-valued variables --- an idea first proposed a decade ago. Other heuristics depend on domain encodings that supply additional information about fluents, such as capacity constraints or other resource-related annotations. A particular challenge to these approaches is in handling interactions between metric fluents that represent exchange, such as the transformation of quantities of raw materials into quantities of processed goods, or trading of money for materials. The usual relaxation of metric fluents is often very poor in these situations, since it does not recognise that resources, once spent, are no longer available to be spent again. We present a heuristic for numeric planning problems building on the propositional relaxed planning graph, but using a mathematical program for numeric reasoning. We define a class of producer--consumer planning problems and demonstrate how the numeric constraints in these can be modelled in a mixed integer program (MIP). This MIP is then combined with a metric Relaxed Planning Graph (RPG) heuristic to produce an integrated hybrid heuristic. The MIP tracks resource use more accurately than the usual relaxation, but relaxes the ordering of actions, while the RPG captures the causal propositional aspects of the problem. We discuss how these two components interact to produce a single unified heuristic and go on to explore how further numeric features of planning problems can be integrated into the MIP. We show that encoding a limited subset of the propositional problem to augment the MIP can yield more accurate guidance, partly by exploiting structure such as propositional landmarks and propositional resources. Our results show that the use of this heuristic enhances scalability on problems where numeric resource interaction is key in finding a solution.