Language Model Post-Training

Follow this topic Ask about this topic

Momentum

23 papers in the last four weeks, up 283% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 132

May 13, 2026cs.LG

Beyond Mode-Seeking RL: Trajectory-Balance Post-Training for Diffusion Language Models

Diffusion language models are a promising alternative to autoregressive models, yet post-training methods for them largely adapt reward-maximizing objectives. We identify a central failure mode in this setting we call trajectory locking: sampled reward-driven updates over-concentrate probability mass onto a narrow set of denoising paths, reducing coverage of alternative correct solutions under repeated sampling. To address this, we propose TraFL (Trajectory Flow baLancing), a trajectory-balance objective that trains the policy toward a reward-tilted target distribution anchored to a frozen reference model. We make this practical for diffusion language models with a diffusion-compatible sequence-level surrogate and a learned prompt-dependent normalization. Across mathematical reasoning and code generation benchmarks, TraFL is the only evaluated post-training method that improves over the base model in every benchmark-length setting, with gains that persist as the sampling budget increases. The improvements transfer to held-out evaluations: TraFL stays above the base model on Minerva Math and is the strongest method on every LiveCodeBench difficulty split.

Saba Ahmadi, Prasanna Parthasarathi, Yufei Cui

Noah’s Ark Lab

RL for Language Model Reasoning RL for Diffusion Models Diffusion Language Models Language Model Post-Training

May 12, 2026cs.CL

TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching

Direct Preference Optimization (DPO) is a widely used RL-free method for aligning language models from pairwise preferences, but it models preferences over full sequences even though generation is driven by per-token decisions. Existing token-level extensions typically decompose a sequence-level Bradley-Terry objective across timesteps, leaving per-prefix (state-wise) optimality implicit. We study how to recover token-level preference optimality using only standard sequence-level pairwise comparisons. We introduce Token-level Bregman Preference Optimization (TBPO), which posits a token-level Bradley-Terry preference model over next-token actions conditioned on the prefix, and derive a Bregman-divergence density-ratio matching objective that generalizes the logistic/DPO loss while preserving the optimal policy induced by the token-level model and maintaining DPO-like simplicity. We introduce two instantiations: TBPO-Q, which explicitly learns a lightweight state baseline, and TBPO-A, which removes the baseline through advantage normalization. Across instruction following, helpfulness/harmlessness, and summarization benchmarks, TBPO improves alignment quality and training stability and increases output diversity relative to strong sequence-level and token-level baselines.

Truong Nguyen, Tien-Phat Nguyen, Linh Ngo Van +3

Hanoi University of Science and Technology · University of Stuttgart · German Research Center for Artificial Intelligence +3

Pairwise Preference Learning LLM Alignment Pairwise Preference Evaluation Pairwise Ranking Density Ratio Estimation Preference Optimization+2

May 12, 2026cs.CL

On Predicting the Post-training Potential of Pre-trained LLMs

The performance of Large Language Models (LLMs) on downstream tasks is fundamentally constrained by the capabilities acquired during pre-training. However, traditional benchmarks like MMLU often fail to reflect a base model's plasticity in complex open-ended scenarios, leading to inefficient model selection. We address this by introducing a new task of predicting post-training potential - forecasting a base model's performance before post-training. We propose RuDE (Rubric-based Discriminative Evaluation), a unified framework that bypasses the generation gap of base models by leveraging response discrimination. Guided by our systematic 4C Taxonomy, RuDE constructs controlled contrastive pairs across diverse domains by fine-grained rubric violations. Extensive experiments demonstrate a correlation greater than 90% with post-training performance. Crucially, validation via Reinforcement Learning (RL) confirms that RuDE effectively identifies high-potential smaller models that outperform larger counterparts, offering a compute-efficient mechanism for foundation model development.

Xiaoyuan Li, Yubo Ma, Kexin Yang +5

University of Science and Technology of China · Alibaba Group · National University of Singapore

LLM Evaluation Language Model Post-Training

May 12, 2026cs.LG

Procedural-skill SFT across capacity tiers: A W-Shaped pre-SFT Trajectory and Regime-Asymmetric Mechanism on 0.8B-4B Qwen3.5 Models

We measure procedural-skill SFT contribution across three Qwen3.5 dense scales (0.8B, 2B, 4B) on a 200-task / 40-skill holdout, with Claude Haiku 4.5 as a frontier reference. The corpus is 353 rows of (task + procedural-skill block, Opus chain-of-thought, judge-pass) demonstrations. Main finding. Under matched-path LLM-only scoring, the SFT-attributable procedural-

Δ

lift is roughly uniform across sizes:

+0.070 / +0.040 / +0.075

at 0.8B / 2B / 4B. Variation in post-SFT

Δ

(

-0.005

,

+0.100

,

+0.065

) is dominated by a W-shaped pre-SFT base trajectory (

-0.075

,

+0.060

,

-0.010

, Haiku-4-5 at

+0.030

): the 5-step procedure hurts 0.8B and 4B, helps 2B, and helps frontier Haiku modestly. SFT works hardest in absolute terms where the base struggles with the procedure -- a regime-asymmetric pattern with a falsifiable prediction at 8B/14B. Methodology. (i) A bench format-compliance artifact: 83.5% of the holdout uses a deterministic ANSWER-line extractor that under-counts free-form-prose conclusions; our LLM-only re-judge reveals it was systematically biased against the curated condition. (ii) A negative-iteration sequence at 0.8B: three well-formed recipe variants cluster post-SFT curated pass-rate within a 2 pp band, constraining the absolute-pass-rate ceiling to base capacity rather than recipe. Cross-family judge validation. GPT-5.4 via OpenRouter on all 7 configurations (2800 paired episodes) agrees on the direction of every per-student finding: Cohen's

κ\geq 0.754

, agreement

\geq 93.25\%

, max headline

Δ

shift

\leq 0.035

pp. Two earlier framings -- "format-only learning at 0.8B" and "SFT contribution shrinks at 4B" -- were path-mismatch artifacts; this paper supersedes both. Single-seed evaluation; threats itemised in the paper.

Igor Strozzi

Applied Mathematics Department Federal University of Rio de Janeiro

Supervised Fine-Tuning LLM Evaluation Language Model Scaling Language Model Post-Training

May 12, 2026cs.CL

YFPO: Yoked Feature Preference Optimization with Neuron-Guided Rewards

Preference optimization has become a widely used post-training paradigm for improving the reasoning abilities of large language models. Existing methods typically learn from preferred and dispreferred responses as external behavioral supervision, while largely ignoring capability-related signals encoded in the model's internal representations. In this work, we study whether such internal signals can provide useful auxiliary supervision for mathematical reasoning. We introduce YFPO (Yoked Feature Preference Optimization), a neuron-guided preference optimization framework that couples response-level preference learning with neuron-level rewards. YFPO first uses AttnLRP to identify math-associated internal features, and then derives an auxiliary reward from the activation margin of these neurons between preferred and dispreferred responses. This reward is combined with the standard preference optimization objective, encouraging the model to align external preferences with internal math-related features. We conduct small-scale experiments on GSM8K with a compact language model. Results show that neuron-guided rewards can influence preference optimization dynamics and yield measurable improvements in several settings, suggesting that internal representations can serve as lightweight and interpretable signals for reasoning-oriented post-training.

Yifan Le

Independent Researcher

Preference Optimization Language Model Post-Training Preference Learning LLM Mathematical Reasoning

May 12, 2026cs.CL

Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy Distillation

On-policy distillation (OPD) has emerged as an efficient post-training paradigm for large language models. However, existing studies largely attribute this advantage to denser and more stable supervision, while the parameter-level mechanisms underlying OPD's efficiency remain poorly understood. In this work, we argue that OPD's efficiency stems from a form of ``foresight'': it establishes a stable update trajectory toward the final model early in training. This foresight manifests in two aspects. First, at the \textbf{Module-Allocation Level}, OPD identifies regions with low marginal utility and concentrates updates on modules that are more critical to reasoning. Second, at the \textbf{Update-Direction Level}, OPD exhibits stronger low-rank concentration, with its dominant subspaces aligning closely with the final update subspace early in training. Building on these findings, we propose \textbf{EffOPD}, a plug-and-play acceleration method that speeds up OPD by adaptively selecting an extrapolation step size and moving along the current update direction. EffOPD requires no additional trainable modules or complex hyperparameter tuning, and achieves an average training acceleration of

3\times

while maintaining comparable final performance. Overall, our findings provide a parameter-dynamics perspective for understanding the efficiency of OPD and offer practical insights for designing more efficient post-training methods for large language models.

Yuchen Cai, Ding Cao, Liang Lin +9

1USTC · 2Tencent · 4HKUST(GZ) +3

Efficient Language Model Training On-Policy Distillation Language Model Post-Training

May 12, 2026cs.LG

From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation

On-policy self-distillation has emerged as a promising paradigm for post-training language models, in which the model conditions on environment feedback to serve as its own teacher, providing dense token-level rewards without external teacher models or step-level annotations. Despite its empirical success, what this reward actually measures and what kind of credit it assigns remain unclear. Under a posterior-compatibility interpretation of feedback conditioning, standard in the implicit-reward literature, we show that the self-distillation token reward is a Bayesian filtering increment whose trajectory sum is exactly the pointwise mutual information between the response and the feedback given the input. This pMI can be raised by input-specific reasoning or by input-generic shortcuts, so we further decompose the teacher log-probability along the input axis. Based on this analysis, we propose CREDIT (Contrastive REward from DIsTillation), which isolates the input-specific component with a batch-contrastive baseline. At the sequence level, CREDIT is a teacher-side surrogate for a contrastive pMI objective that also penalizes responses remaining likely under unrelated inputs. Across coding, scientific reasoning, and tool-use benchmarks on two model families, CREDIT delivers the strongest aggregate performance at negligible additional compute.

Guobin Shen, Lei Huang, Xiang Cheng +4

Xiaohongshu Inc. · Institute of Automation, Chinese Academy of Sciences

Reinforcement Learning On-Policy Self-Distillation Token-Level Credit Assignment Language Model Post-Training Contrastive RL Mutual Information

May 11, 2026cs.CL

Annotations Mitigate Post-Training Mode Collapse

Post-training (via supervised fine-tuning) improves instruction-following, but often induces semantic mode collapse by biasing models toward low-entropy fine-tuning data at the expense of the high-entropy pretraining distribution. Crucially, we find this trade-off worsens with scale. To close this semantic diversity gap, we propose annotation-anchored training, a principled method that enables models to adopt the preference-following behaviors of post-training without sacrificing the inherent diversity of pretraining. Our approach is simple: we pretrain on documents paired with semantic annotations, inducing a rich annotation distribution that reflects the full breadth of pretraining data, and we preserve this distribution during post-training. This lets us sample diverse annotations at inference time and use them as anchors to guide generation, effectively transferring pretraining's semantic richness into post-trained models. We find that models trained with annotation-anchored training can attain

6 \times

less diversity collapse than models trained with SFT, and improve with scale.

Jacob Mitchell Springer, Madhu Advani, Lukas Aichberger +7

Carnegie Mellon University · Apple · Johannes Kepler University Linz

Supervised Fine-Tuning Diverse Text Generation Language Model Post-Training Mode Collapse

May 9, 2026cs.LG

The Extrapolation Cliff in On-Policy Distillation of Near-Deterministic Structured Outputs

On-policy distillation (OPD) is widely used for LLM post-training. When pushed with a reward-extrapolation coefficient lambda > 1, the student can lift past the teacher in domain, but past a threshold lambda* the same step violates the output contract on structured-output tasks. In a single-position Bernoulli reduction, we derive a closed-form base-relative clip-safety threshold lambda*(p,b,c) determined by three measurable quantities: the teacher modal probability, the warm-start mass, and the importance-sampling clip strength. Above lambda*, the extrapolated fixed point exits the clip-safe region, changing training from format-preserving to format-collapsing. We extend the rule to calibrated K-ary listwise JSON tasks where a single binding equivalence class dominates the output contract and SFT retains parse headroom. On Amazon Fashion, three pre-registered tests--a fine-grid cliff interval, a budget-extension test, and a small-clip cross-prediction--fall within their locked prediction windows, with the small-clip value matching the closed-form prediction below grid resolution. Operating just below lambda*, ListOPD brings a 1.7B Qwen3 student to in-domain parity with an 8B-SFT baseline at one-fifth the parameters. The gain is driven primarily by format adherence: NDCG@1 on parsed outputs remains flat across lambda, while parse validity sharply changes at the predicted boundary. The cliff diagnostic is rubric-independent, whereas the parity claim uses a Gemini-graded rubric and inherits that evaluator's exposure.

Xin Li, Hao Jiang, Annan Wang +2

Nanyang Technological University, Singapore

On-Policy Distillation Structured Output Generation Language Model Post-Training Teacher-Student Learning

May 8, 2026cs.CL

Post-training makes large language models less human-like

Large language models (LLMs) are increasingly used as surrogates for human participants, but it remains unclear which models best capture human behavior and why. To address this, we introduce Psych-201, a novel dataset that enables us to measure behavioral alignment at scale. We find that post-training -- the stage that turns base models into useful assistants -- consistently reduces alignment with human behavior across model families, sizes, and objectives. Moreover, this misalignment widens in newer model generations even as base models continue to improve. Finally, we find that persona-induction -- a popular technique for eliciting human-like behavior by conditioning models on participant-specific information -- does not improve predictions at the level of individuals. Taken together, our results suggest that the very processes that are currently employed to turn LLMs into useful assistants also make them less accurate models of human behavior.

Marcel Binz, Elif Akata, Abdullah Almaatouq +76

LLM Evaluation LLM Alignment Human Behavior Prediction Language Model Post-Training Persona-Conditioned Generation

May 8, 2026cs.LG

The Convergence Gap: Instruction-Tuned Language Models Stabilize Later in the Forward Pass

Final outputs hide when a checkpoint commits to its next-token prediction. We introduce the convergence gap, a model-diffing diagnostic that decodes each layer's next-token distribution and measures its distance to the model's own final distribution. Across six paired pretrained and instruction-tuned checkpoints in native prompting regimes, instruction-tuned checkpoints remain farther from their final predictions later into the stack. The effect persists under endpoint-matched raw and tuned readouts, endpoint-free same-history checks, and fixed-history template replay. Matched-prefix interventions identify late MLP windows as the largest tested leverage point: late IT grafts into PT hosts increase late KL by +0.34 nats, while PT-late swaps into IT hosts reduce it by -0.51 nats; matched random late perturbations give only +0.003 versus +0.327 for the true late graft. A preselected Gemma case study provides behavior-facing plausibility for the same late swap, without serving as a benchmark claim. These results identify a robust predictiondynamics signature of post-training: released instruction-following checkpoints tend to settle later, and late MLP computation is the strongest tested bidirectional handle on that delay under matched histories.

Yifan Zhou

University of California, Los Angeles

LLM Interpretability Next-Token Prediction Language Model Post-Training Causal Interventions in Language Models

May 8, 2026cs.LG

Dr. Post-Training: A Data Regularization Perspective on LLM Post-Training

Data selection methods address a critical challenge in LLM post-training: effectively leveraging scarce, high-fidelity target data alongside abundant but imperfectly aligned general training data. In this work, we move beyond the data-selection framing and introduce Dr. Post-Training (Data-Regularized Post-Training), a novel framework that reconceptualizes general training data as a data-induced regularizer that prevents overfitting to the scarce target objective, rather than serving as a pool for selection. Specifically, our framework proposes that at each training step, construct a feasible set of model update directions using the general training data, and project the model update direction specified by the scarce target data onto that feasible set. Standard training and existing data selection methods arise as special cases with different choices of the data-induced regularizer, and these methods correspond to different points on a bias--variance spectrum with different regularization strength. Building on this view, we propose a family of methods offering a richer design space and more flexible bias--variance tradeoffs. For practical LLM-scale use, we introduce careful system optimizations that realize these methods with minimal overhead. Extensive experiments across SFT, RLHF, and RLVR show that our methods consistently outperform state-of-the-art data selection baselines, and system benchmarks confirm their efficiency.

Pingbang Hu, Xueshen Liu, Z. Morley Mao +1

University of Illinois Urbana–Champaign · University of Michigan

Supervised Fine-Tuning Training Data Selection RL Post-Training LLM Post-Training Language Model Post-Training

May 7, 2026cs.AI

Post-Reasoning: Improving the Performance of Non-Thinking Models at No Cost

As the widespread adoption of Large Language Models (LLMs) accelerates, token consumption from intermediate reasoning traces increasingly contributes to inference latency and operational cost. Recent studies suggest that many real-world tasks require little to no explicit reasoning, with additional reasoning sometimes even degrading performance. In this work, we propose Post-Reasoning, a simple yet effective approach that improves instruction-tuned models by conditioning them to justify their answers after generating the final response. By design, it enables the final answer to be obtained without additional latency or token cost, while still improving performance through simple instruction augmentation. We evaluate Post-Reasoning across 117 model--benchmark settings spanning 13 open and proprietary models, 4 model families, and 9 diverse reasoning and knowledge-intensive benchmarks, including AMC, HMMT, GSM8K, GPQA, MMLU-Pro, and BIG-Bench Hard. Post-Reasoning improves performance in over 88.19% of evaluated settings, achieving a mean relative improvements of 17.37%. Furthermore, we propose supervised post-reason tuning, which further improves performance in over 91.11% of evaluated settings, and exceeds the prompt-based post-reasoning baseline by an average of 8.01%, demonstrating that post-reasoning can be effectively internalized through training. Ultimately, Post-Reasoning establishes a new performance ceiling for direct-answer capabilities.

Richmond Sin Jing Xuan, Rishabh Bhardwaj, Soujanya Poria

Nanyang Technological University

LLM Prompting Language Model Post-Training LLM Reasoning

May 6, 2026cs.CL

RLearner-LLM: Balancing Logical Grounding and Fluency in Large Language Models via Hybrid Direct Preference Optimization

Direct Preference Optimization (DPO), the efficient alternative to PPO-based RLHF, falls short on knowledge-intensive generation: standard preference signals from human annotators or LLM judges exhibit a systematic verbosity bias that rewards fluency over logical correctness. This blindspot leaves a logical alignment gap -- SFT models reach NLI entailment of only 0.05-0.22 despite producing fluent text. We propose RLearner-LLM with Hybrid-DPO: an automated preference pipeline that fuses a DeBERTa-v3 NLI signal with a verifier LLM score, removing human annotation while overcoming the "alignment tax" of single-signal optimization. Evaluated across five academic domains (Biology, Medicine, Law) with three base architectures (LLaMA-2-13B, Qwen3-8B, Gemma 4 E4B-it), RLearner-LLM yields up to 6x NLI improvement over SFT, with NLI gains in 11 of 15 cells and consistent answer-coverage gains. On Gemma 4 E4B-it (4.5B effective params), Hybrid-DPO lifts NLI in four of five domains (+11.9% to +2.4x) with faster inference across all five, scaling down to compact base models without losing the alignment-tax mitigation. Our Qwen3-8B RLearner-LLM wins 95% of pairwise comparisons against its own SFT baseline; GPT-4o-mini in turn wins 95% against our concise output -- alongside the 69% win the same judge gives a verbose SFT over our DPO model, this replicates verbosity bias on a frontier comparator and motivates logic-aware metrics (NLI, ACR) over LLM-as-a-judge for knowledge-intensive generation.

Qiming Bao, Juho Leinonen, Paul Denny +1

University of Auckland · Aalto University

LLM Grounding LLM Alignment Direct Preference Optimization Language Model Post-Training

May 5, 2026cs.AI

EvoLM: Self-Evolving Language Models through Co-Evolved Discriminative Rubrics

Language models encode substantial evaluative knowledge from pretraining, yet current post-training methods rely on external supervision (human annotations, proprietary models, or scalar reward models) to produce reward signals. Each imposes a ceiling. Human judgment cannot supervise capabilities beyond its own, proprietary APIs create dependencies, and verifiable rewards cover only domains with ground-truth answers. Self-improvement from a model's own evaluative capacity is a reward source that scales with the model itself, yet remains largely untapped by current methods. We introduce EVOLM, a post-training method that structures this capacity into explicit discriminative rubrics and uses them as training signal. EVOLM trains two capabilities within a single language model in alternation: (1) a rubric generator producing instance-specific evaluation criteria optimized for discriminative utility, which maximizes a small frozen judge's ability to distinguish preferred from dispreferred responses; and (2) a policy trained using those rubric-conditioned scores as reward. All preference signals are constructed from the policy's own outputs via temporal contrast with earlier checkpoints, requiring no human annotation or external supervision. EVOLM trains a Qwen3-8B model to generate rubrics that outperform GPT-4.1 on RewardBench-2 by 25.7%. The co-trained policy achieves 69.3% average on the OLMo3-Adapt suite, outperforming policies trained with GPT-4.1 prompted rubrics by 3.9% and with the state-of-the-art 8B reward model SkyWork-RM by 16%. Overall, EVOLM demonstrates that structuring a model's evaluative capacity into co-evolving discriminative rubrics enables self-improvement without external supervision.

Shuyue Stella Li, Rui Xin, Teng Xiao +8

University of Washington · 2Allen Institute for AI · University of Pennsylvania

RL for Language Models LLM Evaluation Language Model Generation Evaluation Self-Training for Language Models Language Model Self-Assessment Language Model Post-Training+1

Apr 30, 2026cs.CL

Assert, don't describe: Linguistic features that shift LLM reasoning about animal welfare

Animal-welfare advocates produce a lot of writing, and increasingly that writing trains the language models that millions of people then ask about animal welfare. Using vocabulary-matched stance-contrast probes on a held-out animal-welfare benchmark, we measure how each of ten linguistic features changes Llama-3.2-1B's preference for pro-animal-welfare reasoning when used as fine-tuning data. Eight of the ten features produce statistically significant shifts. Seven move the model toward stronger pro-animal-welfare reasoning: assertive certainty, explicit moral vocabulary, emotion words, evaluative claims, narrative structure, depicted harm severity, and immediate temporal framing. Two move it the other way: hedged language and concrete sensory description both dilute the pro-animal-welfare stance. First-person perspective has no statistically significant effect. The practical recommendation for anyone writing animal-welfare text that may end up in LLM training corpora: assert a position rather than describe a scene neutrally. The features that shift the model are the ones that make the writer's position explicit; the features that dilute it hold animal-welfare content but withhold stance.

Jasmine Brazilek, Harper Dunn

Compassion Aligned Machine Learning (CaML) · Independent researcher

Moral Reasoning in Language Models Language Model Post-Training LLM Training LLM Reasoning

Apr 30, 2026cs.CL

Helpfulness Hurts: Domain-Dependent Degradation of Mid-Trained Compassion Values Under Post-Training

Standard post-training pipelines apply supervised fine-tuning (SFT) and reinforcement learning (RL) to make language models helpful, but these processes may inadvertently degrade values instilled during pre-training. We investigate whether the domain of post-training data differentially affects the retention of animal compassion values in a Llama 3.1 8B model mid-trained on compassion-oriented synthetic data, using both SFT (helpfulness via Dolly-15k vs. coding via Magicoder-110K) and GRPO (helpfulness via RLHFlow vs. coding via Magicoder), evaluated on the ANIMA 2.2 benchmark and MORU benchmark (Moral Reasoning Under Uncertainty). Helpfulness training significantly degrades animal compassion relative to coding training on ANIMA (SFT: 35.7% vs. 65.2%; GRPO: 18.7% vs. 32.0%), replicating across two independent helpfulness datasets and two training paradigms. On English MORU items, helpfulness training degrades general moral reasoning by 25.5 percentage points (46.4% vs. 71.9%), a striking gap that rivals the compassion effect in magnitude. However, this effect does not transfer cross-lingually: on the multilingual MORU benchmark, the domain effect disappears (SFT: 52.3% vs. 51.2%). In contrast, the animal compassion effect transfers consistently across languages, with Magicoder's ANIMA percentage-point gain over the base model 4.5 times larger on non-English items than English items. This divergence suggests that values instilled through mid-training are encoded more deeply and cross-lingually than reasoning improvements from domain-specific post-training. These results suggest that, for labs building on value-laden mid-training, coding-domain post-training may better preserve mid-trained values than helpfulness post-training without harming general reasoning capabilities.

Jasmine Brazilek, Juliana Seawell

†Compassion Aligned Machine Learning (CaML)

Moral Reasoning in Language Models RL Post-Training Language Model Post-Training Moral Reasoning

Apr 30, 2026cs.CL

Instruction-Guided Poetry Generation in Arabic and Its Dialects

Poetry has long been a central art form for Arabic speakers, serving as a powerful medium of expression and cultural identity. While modern Arabic speakers continue to value poetry, existing research on Arabic poetry within Large Language Models (LLMs) has primarily focused on analysis tasks such as interpretation or metadata prediction, e.g., rhyme schemes and titles. In contrast, our work addresses the practical aspect of poetry creation in Arabic by introducing controllable generation capabilities to assist users in writing poetry. Specifically, we present a large-scale, carefully curated instruction-based dataset in Modern Standard Arabic (MSA) and various Arabic dialects. This dataset enables tasks such as writing, revising, and continuing poems based on predefined criteria, including style and rhyme, as well as performing poetry analysis. Our experiments show that fine-tuning LLMs on this dataset yields models that can effectively generate poetry that is aligned with user requirements, based on both automated metrics and human evaluation with native Arabic speakers. The data and the code are available at https://github.com/mbzuai-nlp/instructpoet-ar

Abdelrahman Sadallah, Kareem Elozeiri, Mervat Abassy +5

Mohamed bin Zayed University of Artificial Intelligence

Arabic NLP Controllable Text Generation Language Model Post-Training Text Generation

Apr 28, 2026cs.CL

Backtranslation Augmented Direct Preference Optimization for Neural Machine Translation

Contemporary neural machine translation (NMT) systems are almost exclusively built by training on supervised parallel data. Despite the tremendous progress achieved, these systems still exhibit persistent translation errors. This paper proposes that a post-training paradigm based on reinforcement learning (RL) can effectively rectify such mistakes. We introduce a novel framework that requires only a general text corpus and an expert translator which can be either human or an AI system to provide iterative feedback. In our experiments, we focus specifically on English-to-German translation as a representative high-resource language pair. Crucially, we implement this RL-based post-training using Direct Preference Optimization (DPO). Applying our DPO-driven framework to the gemma3-1b model yields a significant improvement in translation quality, elevating it's COMET score from 0.703 to 0.747 on the English to German task. The results demonstrate that DPO offers an efficient and stable pathway for enhancing pre-trained NMT models through preference-based post-training.

Mehrdad Ghassabi, Spehr Rajabi, Hamidreza Baradaran Kashani +3

Faculty of Computer Engineering University of Isfahan Isfahan, Iran · School of Computer Science University of Windsor Windsor, Canada

Language Model Post-Training Machine Translation

Apr 25, 2026cs.CL

$\mathcal{S}^2$ IT: Stepwise Syntax Integration Tuning for Large Language Models in Aspect Sentiment Quad Prediction

Aspect Sentiment Quad Prediction (ASQP) has seen significant advancements, largely driven by the powerful semantic understanding and generative capabilities of large language models (LLMs). However, while syntactic structure information has been proven effective in previous extractive paradigms, it remains underutilized in the generative paradigm of LLMs due to their limited reasoning capabilities. In this paper, we propose S^2IT, a novel Stepwise Syntax Integration Tuning framework that progressively integrates syntactic structure knowledge into LLMs through a multi-step tuning process. The training process is divided into three steps. S^2IT decomposes the quadruple generation task into two stages: 1) Global Syntax-guided Extraction and 2) Local Syntax-guided Classification, integrating both global and local syntactic structure information. Finally, Fine-grained Structural Tuning enhances the model's understanding of syntactic structures through the prediction of element links and node classification. Experiments demonstrate that S^2IT significantly improves state-of-the-art performance across multiple datasets. Our implementation will be open-sourced at https://github.com/DMIRLAB-Group/S2IT.

Bingfeng Chen, Chenjie Qiu, Yifeng Xie +3

School of Computer Science, Guangdong University of Technology · Guangdong Laboratory of Artificial Intelligence and Digital Economy (SZ) · Peng Cheng Laboratory +1

LLM Fine-Tuning Aspect-Based Sentiment Analysis Language Model Post-Training

Apr 24, 2026cs.CL

Thinking Without Words: Efficient Latent Reasoning with Abstract Chain-of-Thought

While long, explicit chains-of-thought (CoT) have proven effective on complex reasoning tasks, they are costly to generate during inference. Non-verbal reasoning methods have emerged with shorter generation lengths by leveraging continuous representations, yet their performance lags behind verbalized CoT. We propose

\textbf{Abstract Chain-of-Thought}

, a discrete latent reasoning post-training mechanism in which the language model produces a short sequence of tokens from a reserved vocabulary in lieu of a natural language CoT, before generating a response. To make previously unseen ''abstract'' tokens useful, we introduce a policy iteration-style warm-up loop that alternates between (i.) bottlenecking from a verbal CoT via masking and performing supervised fine-tuning, and (ii.) self-distillation by training the model to generate abstract tokens from the prompt alone via constrained decoding with the codebook. After warm-up, we optimize the generation of abstract sequences with warm-started reinforcement learning under constrained decoding. Abstract-CoT achieves up to

11.6\times

fewer reasoning tokens while demonstrating comparable performance across mathematical reasoning, instruction-following, and multi-hop reasoning, and generalizes across language model families. We also find an emergent power law distribution over the abstract vocabulary, akin to those seen in natural language, that evolves across the training phases. Our findings highlight the potential for post-training latent reasoning mechanisms that enable efficient inference through a learned abstract reasoning language.

Keshav Ramji, Tahira Naseem, Ramón Fernandez Astudillo

IBM Research AI

Efficient Language Model Reasoning Implicit Reasoning in Language Models Efficient Language Model Inference Language Model Post-Training LLM Reasoning Latent Reasoning+1

Apr 23, 2026cs.CL

On Reasoning Behind Next Occupation Recommendation

In this work, we develop a novel reasoning approach to enhance the performance of large language models (LLMs) in future occupation prediction. In this approach, a reason generator first derives a ``reason'' for a user using his/her past education and career history. The reason summarizes the user's preference and is used as the input of an occupation predictor to recommend the user's next occupation. This two-step occupation prediction approach is, however, non-trivial as LLMs are not aligned with career paths or the unobserved reasons behind each occupation decision. We therefore propose to fine-tune LLMs improving their reasoning and occupation prediction performance. We first derive high-quality oracle reasons, as measured by factuality, coherence and utility criteria, using a LLM-as-a-Judge. These oracle reasons are then used to fine-tune small LLMs to perform reason generation and next occupation prediction. Our extensive experiments show that: (a) our approach effectively enhances LLM's accuracy in next occupation prediction making them comparable to fully supervised methods and outperforming unsupervised methods; (b) a single LLM fine-tuned to perform reason generation and occupation prediction outperforms two LLMs fine-tuned to perform the tasks separately; and (c) the next occupation prediction accuracy depends on the quality of generated reasons. Our code is available at https://github.com/Sarasarahhhhh/job_prediction.

Shan Dong, Palakorn Achananuparp, Hieu Hien Mai +3

Singapore Management University · Columbia University

Large Language Model-Based Recommendation Language Model Post-Training LLM Reasoning

Apr 22, 2026cs.CL

Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech

Simultaneous speech translation (SST) generates translations while receiving partial speech input. Recent advances show that large language models (LLMs) can substantially improve SST quality, but at the cost of high computational overhead. To reduce this cost, prior work reformulates SST as a multi-turn dialogue task, enabling full reuse of the LLM's key-value (KV) cache and eliminating redundant feature recomputation. However, this approach relies on supervised fine-tuning (SFT) data in dialogue form, for which few human annotations exist, and existing synthesis methods cannot guarantee data quality. In this work, we propose a Hierarchical Policy Optimization (HPO) approach that post-train models trained on imperfect SFT data. We introduce a hierarchical reward that balances translation quality and latency objectives. Experiments on English to Chinese/German/Japanese demonstrate improvements of over +7 COMET score and +1.25 MetricX score at a latency of 1.5 seconds. Comprehensive ablation studies further validate the effectiveness of different quality rewards, hierarchical reward formulations, and segmentation strategies. Code can be found here https://github.com/owaski/HPO

Siqi Ouyang, Shuoyang Ding, Oleksii Hrinchuk +4

1Carnegie Mellon University · 2NVIDIA

RL for Language Models Speech Translation Simultaneous Speech Translation Language Model Post-Training

Apr 21, 2026cs.LG

Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training

Tabular language models can generate synthetic tables by modeling rows as token sequences, but they are typically trained once with supervised fine-tuning and then used as static synthesizers. This is limiting because next-token likelihood does not directly optimize the distributional, utility, and indistinguishability properties used to evaluate synthetic data. We study iterative reward-guided post-training for tabular language models through a generate--score--align protocol, where a generator samples synthetic rows, a task-specified reward ranks them, and the model is updated relative to a fixed supervised reference. Within this protocol, we propose \textbf{TabGRAA} (\textbf{Tab}ular \textbf{G}roup-\textbf{R}elative \textbf{A}dvantage \textbf{A}lignment), a group-relative alignment method that compares high- and low-reward generated groups using group-averaged policy/reference log-ratios rather than one-to-one preference pairs. Across five mixed-type benchmarks, TabGRAA improves a GReaT backbone beyond additional supervised fine-tuning and achieves the strongest average trade-off among adapted DPO, KTO, and NPO baselines on fidelity and downstream utility, while maintaining empirical privacy diagnostics near the supervised baseline. Ablations show that the gains depend on meaningful reward ranking and stable group-level updates rather than extra training alone. Reward-substitution and scorer-separation studies further show that the post-training loop can use both classifier-based and classifier-free rewards, and that proper scorer separation is important for preserving the fidelity--utility--privacy trade-off. These results position TabGRAA as a self-improving post-training method for tabular language-model generators, complementary to strong static tabular synthesizers.

Yunbo Long, Tejumade Afonja, Guangya Hao +2

Department of Engineering, University of Cambridge · CISPA Helmholtz Center for Information Security, Saarbrücken, Germany · The Alan Turing Institute, London

Tabular Foundation Models Group Relative Policy Optimization Synthetic Tabular Data Generation Language Model Post-Training Language Model Self-Improvement

Apr 20, 2026cs.LG

Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts

Extending a fully post-trained language model with new domain capabilities is fundamentally limited by monolithic training paradigms: retraining from scratch is expensive and scales poorly, while continued training often degrades existing capabilities. We present BAR (Branch-Adapt-Route), which trains independent domain experts, each through its own mid-training, supervised finetuning, and reinforcement learning pipeline, and composes them via a Mixture-of-Experts architecture with lightweight router training. Unlike retraining approaches that mix all domains and require full reprocessing for any update (with cost scaling quadratically), BAR enables updating individual experts independently with linear cost scaling and no degradation to existing domains. At the 7B scale, with experts for math, code, tool use, and safety, BAR achieves an overall score of 49.1 (averaged across 7 evaluation categories), matching or exceeding re-training baselines (47.8 without mid-training, 50.5 with). We further show that modular training provides a structural advantage: by isolating each domain, it avoids the catastrophic forgetting that occurs when late-stage RL degrades capabilities from earlier training stages, while significantly reducing the cost and complexity of updating or adding a domain. Together, these results suggest that decoupled, expert-based training is a scalable alternative to monolithic retraining for extending language models.

Jacob Morrison, Sanjay Adhikesaven, Akshita Bhagia +3

University of Washington · Allen Institute for AI · University of California, Berkeley

Continual Learning for LLMs Mixture-of-Experts Language Models Language Model Post-Training

Apr 20, 2026cs.LG

Semantic Step Prediction: Multi-Step Latent Forecasting in LLM Reasoning Trajectories via Step Sampling

Semantic Tube Prediction (STP) leverages representation geometric to regularize LLM hidden-state trajectories toward locally linear geodesics during fine-tuning, thereby greatly improving data efficiency. The original STP recipe samples random token sub-spans, which is compatible with the base large language model (LLM) training architecture. Inspired by STP, we are interested to investigate whether the sampling position can further enhance the semantic structure of multi-step reasoning, and hence affect its geometric impact. We applied STP at consecutive semantic reasoning step boundaries and achieved 168x more accurate multi-step latent prediction than frozen baselines on ProcessBench (3,400 samples), compared to only 4x for the random-token STP. Probing the latent manifold with a learned non-linear predictor reveals that STP-shaped trajectories are smooth curves, not straight lines: a 3-layer MLP reduces prediction error by a further 3-12x over linear extrapolation on step-boundary models. Removing the language modeling loss yields trajectories that are 2x more MLP-predictable than the combined loss, revealing a tradeoff between generation quality and geometric purity. Our results identify sampling position as the critical variable in geometric regularization and establish multi-step latent prediction MSE as a new evaluation metric for this class of methods.

Yidi Yuan

Home Team Science and Technology Agency

Neural Representation Geometry Language Model Post-Training

Apr 20, 2026cs.CL

Post-training is (Massive) Supervised Learning

The prevailing paradigm for training LLMs has evolved to rely on a massive post-training phase consisting of SFT and RL. In this position paper, we argue that this methodology effectively marks a reversion to the

pre-train then fine-tune'' approach of the BERT era, explicitly tailoring models to the desired behaviors and specific benchmarks on which they are evaluated. We begin with a historical overview of LLMs, describing the different phases of the LLM evolution. We argue that the current landscape is remarkably similar to the early days of LLMs, where task performance heavily relied on fitting the models to in-distribution datasets. To empirically demonstrate this, we compare pre-trained models to randomly initialized ones, by fine-tuning both variants on modern reasoning datasets and evaluating them on competitive math and code benchmarks. We show that models post-trained from scratch yield highly non-trivial performance. Our findings suggest that current post-training methodologies function primarily as a distribution-fitting mechanism. We finish by positing that developing generally capable models and systems requires moving beyond extensive post-training for predefined behaviors, shifting instead toward training procedures where models

learn how to learn''.

Michael Hassid, Yossi Adi, Roy Schwartz

1FAIR, Meta AI · 2The Hebrew University of Jerusalem

Supervised Fine-Tuning Language Model Post-Training

Apr 19, 2026cs.CL

PoliLegalLM: A Technical Report on a Large Language Model for Political and Legal Affairs

Large language models (LLMs) have achieved remarkable success in general-domain tasks, yet their direct application to the legal domain remains challenging due to hallucinated legal citations, incomplete knowledge coverage, and weak structured reasoning. To address these issues, we propose PoliLegalLM, a domain-specific large language model tailored for political and legal applications. Our approach adopts a unified training framework that integrates continued pretraining, progressive supervised fine-tuning, and preference-based reinforcement learning to jointly enhance legal knowledge grounding, task alignment, and reasoning capability. We construct a large-scale, high-quality legal corpus and design a structured post-training pipeline, enabling the model to effectively learn domain-specific knowledge and adapt to diverse legal tasks. We evaluate PoliLegalLM on three representative benchmarks, including LawBench, LexEval, and a real-world dataset, PoliLegal. Experimental results demonstrate that PoliLegalLM achieves strong and consistent performance, outperforming competitive models of similar scale and remaining highly competitive with significantly larger models, while achieving the best results on real-world legal scenarios. These results highlight the effectiveness of our training paradigm and the practical value of domain-specific LLMs for real-world legal applications.

Yuting Huang, Yinghao Hu, Qian Xiao +7

Zhejiang University, Hangzhou, China · Tongyi Lab, Alibaba Group, Hangzhou, China · Zhejiang Zhengfa Xinxi Guanli Zhongxin, Hangzhou, China

LLM Grounding Language Model Post-Training LLM Training Legal Reasoning

Apr 19, 2026cs.CL

OPSDL: On-Policy Self-Distillation for Long-Context Language Models

Extending the effective context length of large language models (LLMs) remains a central challenge for real-world applications. While recent post-training methods have made progress in long-context scaling, they either rely on high-quality supervision data or sparse sequence-level rewards, leading to unstable and inefficient optimization. We propose OPSDL, an On-Policy Self-Distillation method for enhancing the Long-context capabilities of LLMs. Unlike other recent self-distillation methods that inject privileged information and rely on the model's in-context learning ability to act as a teacher, OPSDL leverages the model's own inherently strong short-context capability as a self-teacher to supervise its own generation in long-context scenarios. The model first generates responses conditioned on the full long-context, then the self-teacher provides per-token supervision signals via point-wise reverse KL divergence under the relevant extracted short-context. This dense token-level signal encourages faithful use of relevant evidence and mitigates hallucinations induced by irrelevant context. We evaluate OPSDL on long-context benchmarks across a range of models from 7B to 32B parameters. Results show consistent and substantial improvements across varying context lengths, outperforming standard post-training approaches such as SFT and DPO with higher sample efficiency. Notably, these gains are achieved without degrading general short-context performance. These findings highlight the effectiveness of OPSDL as a scalable and stable approach for long-context learning.

Xinsen Zhang, Zhenkai Ding, Tianjun Pan +4

Baidu Inc.

Long-Context Language Modeling On-Policy Self-Distillation Language Model Distillation Language Model Post-Training

Apr 19, 2026cs.CL

Jupiter-N Technical Report

We present Jupiter-N, a hybrid reasoning model post-trained from Nemotron 3 Super, a fully open-source 120 billion parameter LLM. We target three objectives: (1) agentic capability via uncertainty-curated trajectories; (2) UK cultural alignment via synthetic data grounded in cultural norms; and (3) Welsh language support via parallel corpora and LLM-translated Welsh conversations. Our data curation strategy carefully preserves the base model's capabilities: using our Forget-Me-Not framework, we mix on-policy synthetic replay with off-policy task data to mitigate catastrophic forgetting, and include a mixture of reasoning and non-reasoning traces to maintain Nemotron's hybrid reasoning ability. Jupiter-N achieves standout gains over Nemotron in Welsh (+18 on ARC-Easy, +5.25 on MMLU-Lite), terminal-use (+9.1 on Terminal Bench 2) and instruction following (+4.4 on IFBench), while retaining the base model capabilities. We frame this work as a reproducible template for sovereign post-training: substituting cultural knowledge, institutional corpora, and target languages produces an equivalent pipeline for any country. All model weights and all post-training datasets are publicly released under open licences.

George Drayson

Locai Labs, University College London

LLM Alignment Cultural Alignment Cultural Knowledge in Language Models Language Model Post-Training