Modelpublic

Codeseys/composer-replication-framework

sourceHugging Facemitupdated 4mo agoView on Hugging Face

0likes

rootresearchnotes

revisionmain master

../

file191108265-mastering-atari-go-chess-and-shogi-by-planning-with-a-learned-model.md6 KBdownload

file201109464-counterfactual-credit-assignment-in-model-free-reinforcement-learning.md5 KBdownload

file221114275-solving-math-word-problems-with-process-and-outcome-based-feedback.md5 KBdownload

file230104104-mastering-diverse-domains-through-world-models.md5 KBdownload

file230406767-raft-reward-ranked-finetuning-for-generative-foundation-model-alignmen.md5 KBdownload

file230510601-tree-of-thoughts-deliberate-problem-solving-with-large-language-models-2.md57 KBdownload

file230510601-tree-of-thoughts-deliberate-problem-solving-with-large-language-models.md5 KBdownload

file230520050-lets-verify-step-by-step.md5 KBdownload

file230616803-would-i-have-gotten-that-reward-long-term-credit-assignment-by-counter.md6 KBdownload

file231004406-language-agent-tree-search-unifies-reasoning-acting-and-planning-in-la-2.md88 KBdownload

file231004406-language-agent-tree-search-unifies-reasoning-acting-and-planning-in-la-3.md88 KBdownload

file231004406-language-agent-tree-search-unifies-reasoning-acting-and-planning-in-la.md5 KBdownload

file231006770-swe-bench-can-language-models-resolve-real-world-github-issues.md5 KBdownload

file231108105-diloco-distributed-low-communication-training-of-language-models.md5 KBdownload

file231108516-llms-cannot-find-reasoning-errors-but-can-correct-them-given-the-error.md5 KBdownload

file231209152-evaluating-augmented-reality-communication-how-can-we-teach-procedural.md5 KBdownload

file240201817-llms-cant-plan-but-can-help-planning-in-llm-modulo-frameworks.md5 KBdownload

file240203300-deepseekmath-pushing-the-limits-of-mathematical-reasoning-in-open-lang.md5 KBdownload

file240211651-learning-from-failure-integrating-negative-examples-when-fine-tuning-l.md5 KBdownload

file240411018-many-shot-in-context-learning.md5 KBdownload

file240515383-generating-code-world-models-with-large-language-models-guided-by-mont.md5 KBdownload

file240612543-phase-controlled-heat-modulation-with-aharonov-bohm-interferometers.md4 KBdownload

file240701476-tree-search-for-language-model-agents.md5 KBdownload

file240806195-mutual-reasoning-makes-smaller-llms-stronger-problem-solvers-2.md65 KBdownload

file240806195-mutual-reasoning-makes-smaller-llms-stronger-problem-solvers.md4 KBdownload

file240919256-hybridflow-a-flexible-and-efficient-rlhf-framework.md6 KBdownload

file241020285-swe-search-enhancing-software-agents-with-monte-carlo-tree-search-and-2.md76 KBdownload

file241020285-swe-search-enhancing-software-agents-with-monte-carlo-tree-search-and.md6 KBdownload

file241108794-llm-based-world-models-can-make-decisions-solely-but-rigorous-evaluati.md5 KBdownload

file241114499-understanding-world-or-predicting-future-a-comprehensive-survey-of-wor.md5 KBdownload

file241221139-training-software-engineering-agents-and-verifiers-with-swe-gym.md4 KBdownload

file250104519-rstar-math-small-llms-can-master-math-reasoning-with-self-evolved-deep.md5 KBdownload

file250104519-sysname-small-llms-can-master-math-reasoning-with-self-evolved-deep-th.md85 KBdownload

file250109136-agentic-retrieval-augmented-generation-a-survey-on-agentic-rag.md6 KBdownload

file250109891-evolving-deeper-llm-thinking.md4 KBdownload

file250112599-kimi-k15-scaling-reinforcement-learning-with-llms.md6 KBdownload

file250118512-streaming-diloco-with-overlapping-communication-towards-a-distributed.md5 KBdownload

file250118639-a-comprehensive-survey-of-the-lean-4-theorem-prover-architecture-appli.md4 KBdownload

file250202047-amasquad-a-benchmark-for-amharic-extractive-question-answering.md4 KBdownload

file250210915-first-passage-times-with-fast-immigration.md4 KBdownload

file250212996-eager-updates-for-overlapped-communication-and-computation-in-diloco.md5 KBdownload

file250218449-swe-rl-advancing-llm-reasoning-via-reinforcement-learning-on-open-soft.md6 KBdownload

file250314391-how-much-do-llms-learn-from-negative-examples.md5 KBdownload

file250314476-dapo-an-open-source-llm-reinforcement-learning-system-at-scale.md5 KBdownload

file250320783-understanding-r1-zero-like-training-a-critical-perspective.md5 KBdownload

file250407164-r2e-gym-procedural-environments-and-hybrid-verifiers-for-scaling-open.md5 KBdownload

file250411343-a-minimalist-approach-to-llm-reasoning-from-rejection-sampling-to-rein.md5 KBdownload

file250415275-stop-summation-min-form-credit-assignment-is-all-process-reward-model.md6 KBdownload

file250421798-swe-smith-scaling-data-for-software-engineering-agents.md5 KBdownload

file250518830-on-the-effect-of-negative-gradient-in-group-relative-deep-reinforcemen.md5 KBdownload

file250613358-socratic-rl-a-novel-framework-for-efficient-knowledge-acquisition-thro.md9 KBdownload

file250613585-minimax-m1-scaling-test-time-compute-efficiently-with-lightning-attent.md7 KBdownload

file250621263-dilocox-a-low-communication-large-scale-training-framework-for-decentr.md5 KBdownload

file250718071-group-sequence-policy-optimization.md4 KBdownload

file250721046-a-survey-of-self-evolving-agents-what-when-how-and-where-to-evolve-on.md6 KBdownload

file250921240-tree-search-for-llm-agent-reinforcement-learning.md5 KBdownload

file251002387-cwm-an-open-weights-llm-for-research-on-code-generation-with-world-mod.md6 KBdownload

file251121654-evilgenie-a-reward-hacking-benchmark.md4 KBdownload

file251218832-from-word-to-world-can-large-language-models-be-implicit-text-based-wo.md5 KBdownload

file251221852-a-comedy-of-estimators-on-kl-regularization-in-rl-training-of-llms.md6 KBdownload

file260103905-current-agents-fail-to-leverage-world-model-as-tool-for-foresight.md5 KBdownload

file260112307-rethinking-the-value-of-multi-agent-workflow-a-strong-single-agent-bas.md5 KBdownload

file260118734-self-distilled-reasoner-on-policy-self-distillation-for-large-language.md5 KBdownload

file260119897-self-distillation-enables-continual-learning.md4 KBdownload

file260120802-reinforcement-learning-via-self-distillation.md5 KBdownload

file260122623-symphony-synergistic-multi-agent-planning-with-heterogeneous-language.md5 KBdownload

file260200994-reasoning-and-tool-use-compete-in-agentic-rlfrom-quantifying-interfere.md5 KBdownload

file260211210-swe-minisandbox-container-free-reinforcement-learning-for-building-sof.md5 KBdownload

file260303195-chain-of-world-world-model-thinking-in-latent-motion.md5 KBdownload

file260324477-composer-2-technical-report.md5 KBdownload

file260402460-single-agent-llms-outperform-multi-agent-systems-on-multi-hop-reasonin.md6 KBdownload

file260407466-cross-tokenizer-llm-distillation-through-a-byte-level-interface.md5 KBdownload

file260412147-evaluating-plan-compliance-in-autonomous-programming-agents.md5 KBdownload

file260415149-llms-gaming-verifiers-rlvr-can-lead-to-reward-hacking.md5 KBdownload

file260423488-do-synthetic-trajectories-reflect-real-reward-hacking-a-systematic-stu.md6 KBdownload

file260505029-the-predictive-causal-gap-an-impossibility-theorem-and-large-scale-neu.md5 KBdownload

file260506840-extracting-search-trees-from-llm-reasoning-traces-reveals-myopic-plann.md5 KBdownload

file260607412-socratic-swe-self-evolving-coding-agents-via-trace-derived-agent-skill-2.md5 KBdownload

file260607412-socratic-swe-self-evolving-coding-agents-via-trace-derived-agent-skill.md11 KBdownload

file3-channel-composed-loss-drgrpo-alphasdpo-betatrace-replay-dpo-the-substrate.md7 KBdownload

filea-technical-report-on-composer-2-cursor.md7 KBdownload

fileadr-006-where-channel-3-trace-replay-dpo-is-hosted-prime-rl-ch13-trl-only-sdpo-m.md4 KBdownload

fileadr-decision-backbone-adr-013-isolated-channel-ladder-prune-vs-train-on-all-prec.md7 KBdownload

fileadversarial-synthesis-agents-dont-spontaneously-use-world-models-for-foresight-m.md9 KBdownload

fileadversarial-synthesis-where-the-composer-25-recipe-is-weak-overclaimed-or-unveri.md9 KBdownload

fileamazon-sagemaker-hyperpod-amazon-sagemaker-ai.md7 KBdownload

fileaws-builder-center-2.md1 KBdownload

filebehind-swe-rebench-infrastructure-to-collect-massive-datasets-of-swe-tasks-and-e.md26 KBdownload

filebugs-scaling-data-for-software-engineering-agents.md160 KBdownload

filebuilding-an-rlhf-training-platform-on-amazon-eks-jark-verl.md6 KBdownload

filechain-of-world-world-model-thinking-in-latent-motion.md71 KBdownload

filechannel-1-drgrpo-base-po-objective-menu-grpodr_grpobnpodapogspocispo-honest-k1-v.md7 KBdownload

filechannel-2-sdpo-self-distillation-generalized_jsd_loss-composer-25-targeted-rl-wi.md8 KBdownload

filechannel-3-multi-teacher-trace-replay-dpo-exact-mechanics-replay_trace-extract_dp.md6 KBdownload

filechoosing-between-amazon-sagemaker-training-jobs-and-amazon-sagemaker-hyperpod-a.md8 KBdownload

filecollator-sdpo-mechanics-hint-injection-ctx_teacher-lock-step-alignment-indices-s.md7 KBdownload

filecomposer-2-technical-report.md73 KBdownload

filecomposer-25-training-method-verified-arxiv-ids-sdftsdpoopsdcomposer2-tech-report.md9 KBdownload

filecompute-and-autoscaling-amazon-eks.md103 KBdownload

filecounter-evidence-heterogeneous-per-node-cross-family-dpo-are-contested-at-equal.md7 KBdownload

filecursor-composer-2-review-benchmarks-pricing-and-the-kimi-k25-controversy-explain.md19 KBdownload

filecursor-composer-25-benchmarks-pricing-full-review.md23 KBdownload

filecursor-composer-25-developer-guide-benchmarks-pricing-lushbinary.md19 KBdownload

filecursor-introduces-composer-25-hacker-news.md72 KBdownload

filecwm-an-open-weights-llm-for-research-on-code-generation-with-world-models.md225 KBdownload

filedapo-an-open-source-llm-reinforcement-learning-system-at-scale.md72 KBdownload

filedeepseekmath-pushing-the-limits-of-mathematical-reasoning-in-open-language-model.md92 KBdownload

filedeepswe-training-a-fully-open-sourced-state-of-the-art-coding-agent-by-scaling-r.md24 KBdownload

filediloco-distributed-low-communication-training-of-language-models.md59 KBdownload

filedivergence-tree-prm-free-counterfactual-process-oracle-that-auto-generates-the-s.md13 KBdownload

fileeks-primary-sagemaker-hybrid-architecture-and-the-minimal-repo-delta.md17 KBdownload

fileevolving-deeper-llm-thinking.md82 KBdownload

fileexploring-expert-failures-improves-llm-agent-tuning-openreview.md4 KBdownload

filefeaturedeletion-task-construction-substrate-inversion-4-gate-validator-hackmonit.md9 KBdownload

filefeaturedeletionenv-runtime-reward-kernel-sandbox-protocol-online-curriculum.md8 KBdownload

filefinal_report_socratic-mcts-swe-worldmodel-8f6dea.md65 KBdownload

fileflat-multi-teacher-to-branching-counterfactual-tree-the-exact-delta-channel-3-vs.md6 KBdownload

filegap-fill-counter-evidence-to-direction-2-aux-next-state-loss-as-necessary.md7 KBdownload

filegithub-lasgroupsdpo-reinforcement-learning-via-self-distillation-sdpo-github.md14 KBdownload

filegithub-novasky-aiskyrl-skyrl-a-modular-full-stack-rl-library-for-llms-github.md8 KBdownload

filegithub-siyan-zhaoopsd-github.md7 KBdownload

filegithub-swe-benchswe-smith-neurips-2025-db-spotlight-scaling-data-for-swe-agents.md5 KBdownload

filegroup-sequence-policy-optimization.md31 KBdownload

filegrpo-trainer-hugging-face.md84 KBdownload

filehow-cursors-composer-25-uses-self-distillation-to-beat-the-frontier-llms-at-codi.md12 KBdownload

fileintroducing-composer-25-cursor.md15 KBdownload

filekurateorg-ai-paper-rankings.md490 Bdownload

filelatent-what-if-deliberation-train-it-aux-next-state-loss-deliberation-token-not.md15 KBdownload

filelayered-hintgenerator-the-sdpo-textual-feedback-machinery-template-raw-error-llm.md7 KBdownload

filelens-synthesis-world-modeling-from-scale-vs-explicit-next-state-training-aux-los.md7 KBdownload

filelocal_sgdpy.md29 KBdownload

filemastering-atari-go-chess-and-shogi-by-planning-with-a-learned-model-nature.md18 KBdownload

filemastering-diverse-domains-through-world-models.md72 KBdownload

fileminimax-m1-scaling-test-time-compute-efficiently-with-lightning-attention.md86 KBdownload

filemockmanager-make_diloco_outer_loop-object-store-sync-drops-into-torchft-diloco-u.md7 KBdownload

filemodalspawnexecutor-working-proof-minimal-delta-to-an-eksexecutor-sagemakerexecut.md8 KBdownload

fileno-gpu-left-behind-unlocking-efficiency-with-co-located-vllm-in-trl.md43 KBdownload

fileprm-vs-outcome-reward-counterfactual-credit-prune-vs-train-on-all-locus.md6 KBdownload

fileprune-vs-train-on-all-empirical-backbone-negatives-help-or-hurt-adversarial-synt.md8 KBdownload

fileprune-vs-train-on-all-type-the-negatives-route-by-signal-committed.md17 KBdownload

filer2e-gym-scaling-open-weights-software-engineering-agents-with-procedural-synthet.md96 KBdownload

filereinforcement-learning-via-self-distillation-2.md148 KBdownload

filereinforcement-learning-via-self-distillation.md151 KBdownload

filereinforcement-learning-with-human-feedback-rlhf-for-llms-with-verl-on-kuberay-ra.md7 KBdownload

filereward-hacking-in-self-improving-code-agents-iclr-2026-rsi-central-adversarial-s.md5 KBdownload

filerl-substrate-trl-vs-verl-vs-prime-rl-altered-model-world-model-rl-critique.md9 KBdownload

filesearch-arxiv-e-print-repository.md3 KBdownload

filesecure-agent-sandboxes-on-eks-gvisor-vs-katafirecracker-per-task-isolation-eks-g.md6 KBdownload

fileself-distilled-reasoner-on-policy-self-distillation-for-large-language-models-2.md53 KBdownload

fileself-distilled-reasoner-on-policy-self-distillation-for-large-language-models.md41 KBdownload

fileself-evolving-mcts-self-distillation-flywheel-conditionally-sound-4-non-negotiab.md18 KBdownload

fileserverless-diloco-substrate-serverlessexecutor-protocol-objectstoreallreduce-aws.md7 KBdownload

filesource-id-verification-adversarial-lens-findings-step-2-width-sweep.md8 KBdownload

filestreaming-diloco-with-overlapping-communication-towards-a-distributed-free-lunch.md126 KBdownload

fileswe-benchswe-smith-datasets-at-hugging-face.md10 KBdownload

fileswe-smith.md2 KBdownload

filethe-predictive-causal-gap-an-impossibility-theorem-and-large-scale-neural-eviden.md17 KBdownload

filetime-slicing-gpus-in-kubernetes-nvidia-gpu-operator.md24 KBdownload

filetrace-ingestion-claude-code-jsonl-to-tracestatetraceexample-input-to-replay-tree.md8 KBdownload

filetraining-software-engineering-agents-and-verifiers-with-swe-gym.md69 KBdownload

fileunderstanding-r1-zero-like-training-a-critical-perspective.md66 KBdownload

fileverifysynthesis-latent-world-model-thinking-lens-chain-of-world-next-state-predi.md9 KBdownload

fileverlreadmemd-at-main-verl-projectverl-github.md20 KBdownload

filewe-ran-composer-25-and-25-fast-across-11-skills-surprisingly-fast-won.md8 KBdownload

filewelcome-to-verls-documentation-verl-documentation-2.md7 KBdownload

filewelcome-to-verls-documentation-verl-documentation.md7 KBdownload

filewiring-the-multi-model-tree-into-the-hint-hook-auto-divergence-feedback-sdpo-sib.md5 KBdownload

Codeseys/composer-replication-framework · main · files are served by the source, never re-hosted here