Optimization
-
Learning When to Trust via Selective Context Preference Optimization
2026-08-06 17:59:58Xian Sun, Wei Chow, Yingshuo Wang, Junhao Liu, Wei Gao, Qing Wu, Lingdong KongAbstract
Language models increasingly condition their answers on external signals, and a single misleading one can turn a correct answer wrong. The obvious remedy, training models to resist such signals, hides a failure mode: a model that ignores all context looks robust yet is useless when the context is worth trusting. We recast the problem as selective trust and introduce MIST, a human-annotated benchmark that renders each reasoning item under four matched conditions (clean, misleading, correct-context, and irrelevant-context), together with SC2W, a paired metric counting how often a misleading signal flips a clean-correct answer to wrong. Across a comprehensive benchmark study, we observe that such a susceptibility is universal. We then propose SCOPE, which mines clean-correct/misleading-wrong failures and optimizes a standard Direct Preference Optimization (DPO) objective over matched preference pairs balanced equally across all four conditions, rather than over misleading items alone. Our approach substantially reduces SC2W on popular open-sourced models while preserving accuracy when the added context is clean, correct, or irrelevant. With this work, we argue that models should be judged on selective trust, not on resistance alone.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06377
PDF
-
HarnessOpt-Bench: Evaluating LLMs at Harness Optimization
2026-08-06 17:21:05Varun Ursekar, Apaar Shanker, Yash Maurya, Shehab Yasser, Vijay S. Kalmath, Veronica Chatrath, Yuan XueAbstract
As LLMs are increasingly deployed within agentic systems, their capabilities depend not only on the model weights but also on the harness: the prompts, tools, control flow, memory, and orchestration code surrounding them. This makes automated harness optimization -- the iterative and evaluation-guided improvement of a harness by an AI system -- both an important route to improving AI systems and a demanding capability for AI systems themselves. Yet the community lacks a common protocol for measuring how well frontier LLMs perform at this task. We introduce HarnessOpt-Bench, a benchmark for end-to-end harness optimization under expensive and stochastic evaluation. An optimizer, an LLM paired with a coding harness, receives a target agent's seed harness, graded evaluation feedback, and a fixed target-evaluation budget. It edits the harness and nominates a final candidate, which is scored by its normalized gain over the seed on a held-out test partition that remains inaccessible throughout search. A trusted execution environment enforces the evaluation boundary, meters target-agent resource use, and preserves candidate versions for audit. We evaluate 5 frontier LLMs as optimizers both under a shared coding harness and under their native harnesses across 4 downstream tasks, over 111 scored runs. Experiment results show that optimizer models separate more than the coding harnesses they act through, native harnesses are not consistently superior, and gains vary substantially across tasks and seed regimes. These results establish harness optimization as a measurable and discriminative capability with large space for improvement.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06301
PDF
-
Continual Learning in Transition
2026-08-06 16:07:26Zhiyan Hou, Dan Zhang, Tao Feng, Liyuan Wang, Wei Li, Xiangzhao Hao, Hongyan An, Junfeng Fang, Haokai Ma, Zhaohui Xu, Haiyun Guo, Jinqiao Wang, Tat-Seng ChuaAbstract
Classical continual learning (CL) has primarily focused on enabling models to update and retain knowledge through parameter-centric mechanisms, e.g., training strategies, architectural designs, and weight adaptation. However, emerging paradigms are reshaping the scope of CL beyond this traditional model adaptation view. For instance, on-policy learning broadens the space of update mechanisms; test-time training extends CL from the training phase to inference; and external harness components such as memory, skill libraries, and interaction protocols extend the evolutionary boundaries of model capabilities far beyond the static parameter space. Collectively, these developments indicate a transition from parameter-centric learning toward system-level adaptation. To characterize this transition, we examine the evolution of continual learning through three dimensions: When, How, and Where learning occurs. The How dimension encompasses off-policy, on-policy, and beyond-gradient optimization mechanics. The When dimension captures evolution across pre-training, post-training, and inference-time stages. The Where dimension delineates updates occurring within internal parameters versus external structural constraints. Anchored by this tri-axial framework, we systematically survey representative methods, trace the ongoing transition of continual learning, and discuss the key challenges, broader implications, and future directions arising from this paradigm shift.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06216
PDF
-
CFGPNet: Cross-Attention-Based Fused Gradient Programmed Network Framework for Multispectral Object Detection
2026-08-06 15:59:13Nima Hatami, Karim Faez, Saeed Sharifian, Hamidreza AmindavarAbstract
RGB--T object detection exploits the complementary strengths of visible and infrared imagery, supporting robust perception in low-light, adverse-weather, and complex multi-scale environments. However, existing methods still suffer from insufficient cross-modal interaction, unstable fusion from modality distribution gaps, and the high computational cost of heavy attention-based architectures. To address these issues, CFGPNet is proposed, a Cross-Attention-Based Fused Gradient Programmed Network framework for multispectral object detection. CFGPNet uses an improved GELAN backbone with RepViT-style re-parameterized blocks to strengthen feature representation while preserving computational efficiency. A Cross Computation Efficient Attention (CrossCEA) module is introduced to enhance cross-modal feature interaction and reduce redundant information transfer between visible and thermal branches. To generate compact and discriminative fused representations, an Attention Selection and Aggregation Fusion (ASAF) network combines dense feature aggregation with selective attention-based emphasis. Moreover, a programmable-gradient auxiliary branch is integrated into each CFGPNet variant to improve gradient delivery and optimization quality. Experiments on five public multispectral benchmarks, FLIR, M3FD, LLVIP, VEDAI, and MFAD, demonstrate that CFGPNet achieves strong and consistent performance across diverse scenes, object scales, and modality balances. In particular, the framework attains 80.7% mAP50 / 45.0% mAP50:95 on FLIR, 89.9% / 63.4% on M3FD, and 97.8% / 68.9% on LLVIP. It also reaches 83.3% / 56.9% on VEDAI and 83.4% / 61.8% on MFAD. These results show that CFGPNet is an effective, practical solution offering useful accuracy--efficiency trade-offs across three model scales. The code, data, and fine-tuned models are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06205
PDF
-
MicroEvo: Knowledge-Guided LLM Sampling for Efficient Microarchitecture Design Space Exploration
2026-08-06 15:43:31Jia Xiong, Runkai Li, Chenxu Niu, Guangyuan Gao, Changwen Xing, Yifan Zhang, Xinlai Wan, Jieran Cui, Chen Bai, Yusheng Hua, Ying Wang, Ming Ling, Xi Wang, Tao XieAbstract
Microarchitecture design space exploration suffers from expansive search spaces and expensive PPA evaluation, leaving only a small simulation budget for design decision-making. Existing methods perform blind search without considering microarchitectural dependencies and fail to learn from the iterative search effectively, leading to wasted evaluations and weak Pareto convergence. In this paper, we propose MicroEvo, a knowledge-guided framework that couples off-the-shelf LLMs with Monte Carlo Tree Search (MCTS) for multi-objective microarchitecture optimization. MicroEvo combines LLM-driven evolutionary operators, a Pareto-aware tree policy that balances Pareto contribution and diversity, an active knowledge accumulation mechanism that extracts and reuses optimization insights, and state-aware directives that adapt the search behavior online. Experiments show that MicroEvo improves Pareto-front quality by up to 36.2% over NSGA-II and achieves 10.6x higher search efficiency, and also demonstrates strong scalability to a complex industrial-scale core. The code repository is available at: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06183
PDF
-
iARCS: Iterative Agentic RL for Controllable 3D Scene Generation
2026-08-06 15:30:21Saugat Adhikari, Ashok Prasad Neupane, Pramish Paudel, Ajad Chhatkuli, Danda Pani PaudelAbstract
Synthetic 3D scene generation is increasingly used as a data source for computer vision and embodied AI, but existing generators often optimize perceptual realism without reliably satisfying task-critical functional constraints. This mismatch limits the usefulness of synthetic data for downstream training, where accessibility, traversability, and spatial rule compliance are often essential. We present iARCS, an iterative agentic reinforcement learning framework that adapts a pretrained scene generator to natural-language task requirements. iARCS uses a two-stage strategy: universal-reward pretraining to improve physical plausibility and layout quality, followed by task-specific fine-tuning with LLM-generated reward programs that are iteratively refined from training feedback. Experiments show improved constraint fidelity on walkability, reachability, and clearance-focused tasks, effective task-specific constraint optimization, and competitive scene diversity. We further show that data generated by iARCS improves a base generator, supporting its value as a practical synthetic data generation tool rather than only a controllable scene editing method.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06161
PDF
-
Contextual Information Policy Optimization for Search Agents
2026-08-06 15:01:29Xingyu Guo, Wei Chen, Linlin Yang, Baochang ZhangAbstract
Search agents extend large language models beyond static parametric memory by enabling them to acquire and use ex ternal evidence during multi-step reasoning. For knowledge intensive tasks involving complex or evolving information, their reliability depends not only on retrieving relevant ev idence but also on using it to guide subsequent reasoning. However, existing methods primarily reward final-answer cor rectness or intermediate progress, without directly assessing whether post-retrieval actions are grounded in the retrieved evidence. This misalignment encourages prior-driven reason ing: agents form conclusions based on internal knowledge and use retrieval mainly to confirm them, resulting in confirma tion bias and inefficient this http URL, we propose Contextual Information Policy Optimization (CIPO), an evidence-oriented reinforcement learning framework that explicitly aligns policy optimization with external evidence use. CIPO assigns dense, turn-level credit to reasoning ac tions influenced by retrieved information, while combining this evidence-use signal with a global outcome reward to pre this http URL,CIPOdiscourages evidence-detached guesses and promotes reasoning trajecto ries in which retrieved facts can guide or revise subsequent reasoning. Importantly, CIPO requires neither human process annotations nor an additional reward model. Extensive exper iments on seven in-domain and out-of-domain benchmarks show that CIPO reduces the prevalence of prior-driven rea soning and achieves excellent performance on most tasks.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06128
PDF
-
Sample-Adaptive Latent Rewards for Uncertainty-Guided Diffusion Post-Training
2026-08-06 14:55:42Rui Li, Yuanzhi Liang, Ke Hao, Ziqiao Weng, Haibin Huang, Chi Zhang, XueLong LiAbstract
Latent reward models can supervise visual diffusion models without decoding intermediate states into pixel space. This makes alignment with human preferences more efficient. However, existing latent reward models output only scalar scores. They do not estimate the uncertainty of each prediction. The generator therefore cannot determine which feedback is reliable. This can drive optimization in the wrong direction and lead to reward hacking. We propose \textsc{SURE}, a unified latent-space framework for image and video diffusion models. It learns reward distributions and directly uses their reliability to guide dense post-training. First, we propose sample-adaptive latent reward model (\textsc{SURE-LRM}). It predicts a Gaussian utility for each noisy latent. Its mean predicts the reward score. Its variance reflect the uncertainty of prediction without human annotation. The learned distribution then guides post-training through uncertainty-guided reward feedback learning (\textsc{SURE-REFL}). This method provides uncertainty-guided dense feedback along the denoising trajectory. At selected transitions, \textsc{SURE-REFL} queries the frozen \textsc{SURE-LRM}. It converts detached variance into reliability weights for samples at the same transition. Each weighted reward is backpropagated only through its local transition. The entire process remains in latent space and requires neither pixel-space decoding nor the full denoising graph. Experiments show that \textsc{SURE-LRM} improves preference prediction over strong baselines. \textsc{SURE-REFL} achieves the sota performance among various metrics and further improves optimization stability. It also achieves the highest VBench quality, semantic, and total scores among the evaluated methods.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06125
PDF
-
Confidence matters: Leveraging Multi-view Geometric Priors for GS-based Reconstruction
2026-08-06 14:50:29Hongyu Zhou, Zorah L\"ahnerAbstract
3D Gaussian splatting (3DGS) has emerged as a widely-used tool for novel view synthesis, offering real-time rendering in a sparse representation. However, the method's reliance on structure-from-motion initialization and photometric optimization can lead to suboptimal geometric reconstruction, particularly for objects with high specularity. In this work, we investigate the integration of geometric priors, in the form of predicted normal and depth maps, into the 3DGS framework to improve the reconstruction quality. We analyze the effect of incorporating these priors into GS-based methods and our evaluation reveals that multi-view predictions, as they are done by the recent visual geometry grounded transformer (VGGT), outperform single-view alternatives. A major factor is the existence of a confidence map for the estimations, which comes as a by-product of multi-view models and which can significantly improve the effectiveness of priors by weighting each prediction appropriately. Extensive experiments on standard benchmarks show consistent improvement in reconstruction quality and significant gains in complex scenes including specular objects.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06117
PDF
-
From Siloed Algorithms to Compliance-First Agentic Platforms: A Multi-Layered Architecture for Hospital AI Systems
2026-08-06 14:44:31Manideep Dhar, Ritwik Singh, Sharat Chandra Kumar ManikondaAbstract
Hospitals are rapidly adopting artificial intelligence for triage, imaging, scheduling etc., yet most deployments remain isolated point solutions locked inside departmental silos, resulting in duplicated effort, hidden risks, and unrealized enterprise value. Despite explosive growth of AI in healthcare market and accelerating investment, an estimated 70-80% of healthcare AI pilots fail to scale, largely due to governance gaps, fragmented data, and missing integration blueprints. This research proposes a hospital-specific, compliance-first, Agentic AI architecture with multiple interoperable layers, extending existing hospital AI platform models with: (i) an Agent Orchestration Layer for multi-agent workflows across clinical, operational, and financial domains, (ii) a Compliance and Policy Layer that centralizes policy-as-code for HIPAA, GDPR, the EU AI Act, DISHA Act, India's DPDP Act, and ISO/IEC security and safety standards, and (iii) a Privacy-Preserving Data Fabric that plugs federated learning, differential privacy, and secure enclaves into real-world Hospital Information Management System (HIMS) flows. Using a synthetic but structurally realistic hospital dataset and an open, ready-to-deploy prototype implementation, this study demonstrates the end-to-end orchestration of triage risk prediction, workflow optimization, and compliance logging, achieving substantial simulated reductions in task turnaround times and manual documentation effort while maintaining policy-guarded data access. The resulting architecture offers hospital leaders a pragmatic blueprint to move from ad hoc tools to a governed, globally compliant, ROI-focused AI platform that can be tailored to on-premise, hybrid and cloud-native deployments.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06112
PDF
-
HERALD: Counterfactual Audits and Minimal Repairs for Proof-of-Retrieval Rewards
2026-08-06 13:15:54Zhuowen Liu, Bohan Cui, YinShang Guo, Yuting Wang, Hao LiAbstract
Search-agent rewards mix answer quality, citation grounding, tool cost, and anti-hacking terms; a high score therefore need not imply that cited evidence was retrieved, and added penalties can cancel. We introduce HERALD, an offline audit that applies exact same-question interventions, separates candidate-visible from oracle information, and enumerates detector contracts before policy optimization. On four Qwen3-8B pools from HotpotQA, 2WikiMultiHopQA, and MuSiQue, $R_0$ rejects search deletion and fake IDs, but a label-free citation-laundering attack succeeds. A complete $2^3$ ablation identifies targeted strengthening of $L$---citing a corpus passage absent from the retrieved evidence---as the observed inclusion-minimal repair: $R[L]$ has zero empirical ASR with a 0.50% one-sided cluster upper bound. The gap persists across pool rules, a visible BM25 attacker, and four models; broader hardening remains vulnerable when the attack removes an oracle support-ID penalty. Under strict 5M-token matched training evaluated on 256 paired questions per benchmark, $R[L]$ meets the EM non-inferiority gate on HotpotQA and 2Wiki but not MuSiQue. Equal-suite citation precision and support recall improve by 2.02 and 1.46 points, unsupported citations fall by 1.69, and laundering attackability falls on 2Wiki and MuSiQue. Natural $L$ is not reduced, and the detector appears in only 18 of 58,368 training trajectories. HERALD thus separates robust scoring, sparse learning signal, and policy transfer.
Abstract (translated)
URL
https://arxiv.org/abs/2608.06012
PDF
-
AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning
2026-08-06 13:00:59Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao, Jinyang Wu, Jie Wu, Zhengzhou Cai, Yueqing Sun, Ziang Ye, Linji Hao, Qi Gu, Xunliang Cai, Yongliang Shen, Yujiu YangAbstract
Reinforcement learning (RL) with verifiable rewards constructs trajectory-level advantage estimates, yet it often fails to credit the few pivotal decisions that determine outcomes in long-horizon, multi-turn agentic tasks. Recent work introduces privileged self-distillation for credit assignment, providing denser supervision, but it remains unclear how such local signals should represent sequential credit. We propose AgentOPSD, a critic-free, recursive method for turn-level credit assignment in agentic reinforcement learning. AgentOPSD aggregates token-level teacher-student log-probability gaps into turn-level evidence and recursively updates a Bayesian belief state in log-odds space. This yields a principled reweighting scheme that converts sparse outcome supervision into turn-level credit signals and identifies pivotal turns through the marginal belief revision between consecutive states. The method is fully compatible with standard policy optimization and requires neither an additional critic nor extra rollouts. We evaluate AgentOPSD on ALFWorld, WebShop, and Search-QA using Qwen2.5 models at two scales (3B and 7B). AgentOPSD outperforms GRPO and strong self-distillation baselines, achieving 89.1% success on ALFWorld with Qwen2.5-7B. Ablation studies attribute the gains to turn-level aggregation and history-dependent recursive belief updates.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05987
PDF
-
Universal Concept Disruption for SAM3 Image Segmentation
2026-08-06 13:00:02Hao Wang, Yuxuan Zhang, Wei YangAbstract
SAM3 extends promptable segmentation from geometry-driven mask prediction to open-vocabulary concept segmentation, where a text-conditioned grounding model decides whether a concept is present and segments all matching instances. While this presence-gated design improves concept-level prediction, its adversarial robustness remains unexplored. In this paper, we introduce Universal Concept Disruption (UCD), the first universal cross-concept adversarial attack tailored to SAM3 image segmentation. UCD learns a single bounded image perturbation from (image, noun-phrase) pairs and attacks SAM3 as an integrated concept-grounding system. It jointly disrupts the text-conditioned input path, maximizes divergence in prompt-shared visual features, suppresses the final presence-gated concept scores, and corrupts the spatial validity of retained masks through area collapse and clean-mask Dice disruption. Across SACo-Gold, LVIS, RefCOCO, PhraseCut, and OpenImages datasets, UCD consistently outperforms all baselines under a matched evaluation protocol, reducing average mask AP from 59.43 to 18.73 and average cgF1 from 50.32 to 20.49. The learned perturbation also transfers to SAM3.1 and to SAM3 video inference without re-optimization, while prompt ensembling, lightweight head fine-tuning, and temporal filtering provide limited recovery.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05983
PDF
-
VLMs for Videogame Data Annotation
2026-08-06 12:20:53Katrin Schmid, Iuri FrosioAbstract
Vision Language Models (VLMs) and Artificial Intelligence (AI) agents have revolutionized how engineers approach complex problems in real-world applications. Their adoption in video games is on the other hand limited by the extreme variability of the synthetic scenarios and their poor compliance with real-world physics. Here we investigate the use of VLMs for annotating video game frame sequences with reward signals, a task with several potential applications including, among others, conditioned training and offline reinforcement learning. We show that VLMs often struggle to answer basic questions on racing video games (although we observed a similar behavior on other game genres) and discuss countermeasures such as VLM output mixing and prompt optimization. We also show how input sequence length, resolution, and question batching affect the annotation quality and its token consumption.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05949
PDF
-
Respect Your Zero-Shot Uncertainty: Conservative Calibration for Test-Time-Adapted Vision-Language Models
2026-08-06 12:11:53Jingyan Jiang, Yaru Sun, Xiao Chen, Jiazhen Huang, Caiting Li, Zhijian He, Yin Chen, Pingting HaoAbstract
Test-time adaptation (TTA) can improve the recognition accuracy of vision-language models under distribution shift, but often degrades calibration, making predictive confidence unreliable for downstream decision-making. Many existing label-free calibration approaches are either coupled to prompt optimization or rely on logit-range statistics that provide only a coarse characterization of the predictive distribution. We show that TTA can increase confidence and reduce entropy even when the top-1 prediction and its correctness remain unchanged, a failure mode we term prediction-preserving sharpening. Across diverse TTA methods and benchmarks, larger entropy reductions relative to paired zero-shot predictions are associated with greater increases in Expected Calibration Error (ECE). On entropy-reduced samples, confidence gains also tend to exceed accuracy gains. Based on these findings, we propose Zero-Shot-Anchored Entropy Calibration (ZAEC), a label-free post-hoc method that uses zero-shot entropy as a sample-specific uncertainty reference. ZAEC selectively restores the zero-shot entropy of sharpened predictions through minimal temperature scaling while leaving all other predictions unchanged. It requires no labeled calibration data or learned parameters and preserves class rankings and classification accuracy. Across five TTA methods and 15 datasets, ZAEC achieves the lowest post-hoc macro-average ECE on ViT-B/16, with consistent gains on RN50.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05945
PDF
-
GSBF: Gaussian Splatting for Environment-Aware Beamforming
2026-08-06 11:22:34Yijie Bian, Wei Guo, Zixin Wang, Shenghui Song, Jun Zhang, Khaled B. LetaiefAbstract
Beamforming plays a key role in multiple-input-multiple-output (MIMO) communication systems. However, conventional beamforming design normally requires accurate instantaneous channel state information (CSI) and iterative optimization, which incur substantial pilot overhead and computational complexity. Recognizing that radio propagation is intrinsically governed by the physical geometry, we develop a 3D Gaussian splatting for environment-aware beamforming (GSBF) pipeline based on multi-modal data, which characterizes the environment through a persistent 3D Gaussian representation. Specifically, GSBF models the environmental scattering response with reciprocity-preserving bidirectional spherical Gaussian (Bi-SG) kernels and performs two-sided electromagnetic rasterization to render an angular propagator map. The rendered map is then aggregated through an over-complete array-manifold dictionary and projected to the constant-modulus beamformers, thereby synthesizing beams directly from the access point (AP) pose and user position without online instantaneous CSI. Simulations demonstrate that GSBF consistently outperforms baselines such as exhaustive beam alignment (EBA) with lower latency.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05896
PDF
-
Coordinated Multi-Robot Disassembly for Makespan Optimization of Large-Scale Assemblies
2026-08-06 09:59:42Niklas Hargus, Andreas Orthey, Marc ToussaintAbstract
Multi-robot task and motion planning for disassembly tasks requires robots to operate in confined workspaces while coordinating their motions with other robots. To tackle this problem, we propose a planning method called coordinated multi-robot disassembly (CoMuDi). CoMuDi coordinates a team of robots for disassembly tasks. The input is a team of robots, an assembly of objects, and a dependency graph. Based on this information, we create compound tasks for pick, place, and exit motions. By propagating temporal constraints, we ensure that each robot can start and end their tasks as early as possible while avoiding collisions with nearby robots. By integrating the space-time RRT* planner (ST-RRT*) into CoMuDi, we ensure that individual tasks minimize arrival time and thereby help us minimize overall makespan. We compare the performance of CoMuDi using both ST-RRT* and RRT* planners with varying time bounds, demonstrating that the combination of CoMuDi and ST-RRT* leads to a higher success rate while minimizing makespan. Finally, we evaluate CoMuDi on six assemblies with up to 49 pieces and up to 9 robots. In those scenarios, we show that CoMuDi returns robot paths that exhibit low idle times, thereby demonstrating that CoMuDi can reliably solve large-scale assemblies.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05830
PDF
-
Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation
2026-08-06 09:19:16Tirth Bhatt, Naren Kumar S, Mayank SinghAbstract
Multilingual text embedding models are commonly adapted using a single training objective across diverse tasks, despite different tasks requiring fundamentally different optimization strategies. We introduce Task-Conditional Flow Matching (TCFM), a multilingual embedding adaptation framework that selectively applies Flow Matching to translation tasks while optimizing retrieval, classification, and pair-classification tasks with objectives better aligned to their learning dynamics. TCFM further combines teacher-guided representation preservation with a three-stage curriculum to enable stable adaptation. Evaluated on the Indic Massive Text Embedding Benchmark, TCFM establishes a new state-of-the-art, consistently improving embedding quality across a diverse set of multilingual tasks and generalizing across embedding model families. We will publicly release the codebase and datasets upon acceptance of the paper.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05785
PDF
-
GROM: Gradient-Free Rapid One-Shot Machine Unlearning
2026-08-06 09:16:53Pawe{\l} Batorski, Przemys{\l}aw Spurek, Paul SwobodaAbstract
Machine unlearning has become a critical capability for safely removing specific, sensitive knowledge from large language models (LLMs). Current state-of-the-art approaches primarily rely on iterative, training-time unlearning via fine-tuning. However, even when utilizing parameter-efficient dimensionality reduction techniques like LoRA, gradient-based optimization remains computationally expensive and lacks explicit analytical formulations. It can also leave the targeted knowledge merely hidden rather than removed, to the point that simply quantizing the unlearned model restores much of what it was supposed to have erased. To resolve this, we propose a novel one-shot unlearning approach, abandoning iterative optimization in favor of a direct, exact analytical solution. We frame the unlearning process as a ridge-regularized least-squares optimization problem, deriving a closed-form additive update for targeted weight matrices. This update forces the selected layer to suppress unwanted content while strictly preserving its behavior on retained data. Computed from gradient-free forward passes alone, with no backpropagation and no iteration to convergence, GROM applies the weight edit in mere seconds, which makes it orders of magnitude faster than traditional fine-tuning. Extensive evaluations demonstrate that GROM achieves state-of-the-art forgetting-utility trade-offs on TOFU-5%, TOFU-10%, MUSE-Books, MUSE-News and WMDP, significantly reducing computational overhead without sacrificing overall model performance. Because the update removes the targeted content from the weights instead of masking it, GROM also withstands the low-bit quantization attack that recovers much of the content a gradient-based baseline had appeared to forget. Our code is publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.05783
PDF
-
DASH: Decoupled Adaptive Surrogate - Acquisition Harness for Automated Bayesian Optimization
2026-08-06 09:03:45Changquan Zhao, Yuxiang Sun, Ruihao Zhu, Cheng Hua, Yulian HeAbstract
Bayesian optimization (BO) relies on a surrogate model and an acquisition function, yet the most suitable choices vary across tasks and optimization stages. Automated Bayesian optimization (AutoBO) addresses this variability by adapting BO components online. However, existing AutoBO methods either adapt one component, leaving the other mismatched and creating a bottleneck, or jointly select surrogate--acquisition pairs under a shared criterion, overlooking their distinct roles: surrogate selection depends on predictive reliability, whereas acquisition adaptation should respond to campaign this http URL this paper, we propose DASH, a Decoupled Adaptive Surrogate--Acquisition Harness for large-language- model (LLM)-enhanced AutoBO. DASH selects surrogates by predictive reliability, uncertainty calibration, and ranking consistency; its two-stage acquisition controller periodically reallocates quotas across acquisition functions, builds a BO shortlist accordingly, and delegates final selection to an LLM. DASH also incorporates an integrated harness, consisting of knowledge-guided warm start and structured memory, to ground optimization in domain knowledge and accumulated feedback. Across four chemical optimization tasks, DASH outperforms the best AutoBO baseline by 12.51% in trajectory-level Acceleration Factor and 5.00% in endpoint Enhancement Factor. Results remain strong across LLM backbones, and ablations verify the complementary contributions of all components. Full-table and behavioral contamination checks find no detectable evidence that direct benchmark memorization or source-cell leakage explains these gains.
Abstract (translated)
URL
https://arxiv.org/abs/2608.00641
PDF
